Method and device for early determination of nematode disease in pine

Abstract

The present invention provides a method and device for early detection of pine nematode disease, which belongs to the field of image processing technology and includes the following steps: obtaining monitoring images of a target pine forest at multiple observation times; extracting feature points from each monitoring image to obtain homonymous points for each monitoring image; performing coordinate correction on all monitoring images based on the homonymous points to determine reference information for each pine tree and patch parameters for each patch; generating evaluation parameters for patches on each pine tree based on the reference information for each pine tree and the patch parameters for each patch; and determining the early stage of nematode disease in the target pine forest based on the evaluation parameters. The present invention provides a method and device for early detection of pine nematode disease, based on high-resolution visible light data of infected trees that changes over time, by tracking the temporal process of patch occurrence and development in the canopy of the pine forest, achieving high-precision early diagnosis of pine wilt disease with a low false alarm rate and high universality, effectively reducing ecological and economic losses.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and device for early determination of nematode disease in pine wood. Background Art

[0002] Pine wilt disease is an epidemic disease of pine wood caused by pine wood nematode fungi. It has the characteristics of rapid spread and strong destructiveness, posing a great threat to forestry ecological security.

[0003] The existing early diagnosis method for pine nematode disease is based on a model built on hyperspectral imaging, and is mainly carried out on single discolored wood.

[0004] Since the above method is limited to the laboratory environment, the screened early diagnosis index and bands do not take into account the interference of other external environments such as diseases and environmental stress, resulting in low accuracy, high false alarm rate, and low universality of the method. Summary of the Invention

[0005] The method and device for early determination of pine wood nematode disease provided by the present invention are used to overcome the defects of the existing technology being limited to laboratory environment, and the screened early diagnosis index and band not taking into account the interference of other external environments such as diseases and environmental stress, resulting in low accuracy, high false alarm rate, and low universality of the method. The method can realize early diagnosis of pine wood nematode disease with high accuracy, low false alarm rate, high universality, and can effectively reduce ecological and economic losses.

[0006] The present invention provides a method for early determination of nematode disease in pine wood, comprising:

[0007] Acquire monitoring images of the target pine forest at multiple observation times;

[0008] Extracting feature points from each monitoring image to obtain points with the same name in each monitoring image;

[0009] Based on the same-name points, coordinate correction is performed on all monitoring images to determine the reference information of each pine tree and the patch parameters of each patch;

[0010] generating evaluation parameters of the patches on each pine tree according to the reference information of each pine tree and the patch parameters of each patch;

[0011] The early stage of nematode disease in the target pine forest is determined based on the evaluation parameters.

[0012] According to a method for early determination of pine nematode disease provided by the present invention, extracting feature points from each monitoring image to obtain the same-name points of each monitoring image includes:

[0013] Extract feature points from each monitoring image to determine the feature points of each monitoring image;

[0014] The feature points of the monitoring images collected at the initial observation time among the multiple observation times are used to match the feature points of each monitoring image to obtain the points with the same name in each monitoring image.

[0015] According to the method for early determination of nematode disease in pine wood provided by the present invention, coordinate correction is performed on all monitoring images based on the same-name points to determine reference information of each pine tree and patch parameters of each patch, including:

[0016] Determine each preset equation using the same-name points of the monitoring image collected at the initial observation time;

[0017] Based on each of the preset equations and all the points of the same name, pixel matching is performed on each monitoring image with the monitoring image collected at the initial observation time to generate a corrected image corresponding to each monitoring image;

[0018] Determine the reference information of each pine tree in each calibrated image, as well as the patch parameters of each patch.

[0019] According to the method for early determination of pine nematode disease provided by the present invention, the reference information includes: the geometric center point and target tangent line of the pine tree canopy; the patch parameters include: patch position and patch area;

[0020] Generating evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch includes:

[0021] Determine, based on the geometric center point of each pine tree canopy in each corrected image and the patch position of each patch in each corrected image, the distance information between each patch and the geometric center point of the pine tree canopy where the patch is located; and determine, based on the patch area of each patch on each pine tree in the corrected image corresponding to two adjacent observation moments and the time difference between the two adjacent observation moments, the patch area parameter;

[0022] Obtaining the relative distance of each patch in the crown of each pine tree according to the distance information of each patch on the crown of each pine tree and the target tangent;

[0023] The patch evaluation parameter of each pine tree in each corrected image is determined according to the patch area parameter and the relative distance of each patch.

[0024] According to a method for determining early stage nematode disease in pine wood provided by the present invention, determining the early stage of nematode disease in pine trees in the target pine forest based on the evaluation parameters includes:

[0025] Determining the patch change degree of the target pine forest at each two adjacent observation moments according to the patch evaluation parameters on all the corrected images;

[0026] According to the degree of all patch changes, the early stage of nematode infection in the target pine forest is determined.

[0027] According to the present invention, a method for early determination of nematode disease in pine wood is provided, wherein the method determines the early infection time of nematode disease in the target pine forest based on the degree of change of all patches, including:

[0028] When the degree of patch change at any moment in the target pine forest is greater than a preset threshold, determining a time period corresponding to the degree of change in the presence of diseased trees infected with nematodes in the target pine forest;

[0029] In the time periods corresponding to the degree of change in the presence of diseased trees infected with nematodes in the target pine forest, the first time period is determined to be the early stage of nematode infection in the target pine forest.

[0030] The present invention also provides a device for early determination of nematode disease in pine wood, comprising:

[0031] An acquisition module is used to obtain monitoring images of the target pine forest at multiple observation times;

[0032] A feature point extraction module is used to extract feature points from each monitoring image to obtain points with the same name in each monitoring image;

[0033] A coordinate correction module, configured to perform coordinate correction on all monitoring images based on the same-name points to determine reference information of each pine tree and patch parameters of each patch;

[0034] a generating module, configured to generate evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch;

[0035] A determination module is used to determine the early stage of nematode disease in the target pine forest based on the evaluation parameters.

[0036] The present invention also provides an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for early determination of nematode disease in pine wood as described above is implemented.

[0037] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for early determination of nematode disease in pine wood as described in any one of the above is implemented.

[0038] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned methods for early determination of nematode disease in pine wood.

[0039] The method and device for early determination of pine wood nematode disease provided by the present invention are based on high-resolution visible light data of infected trees with time-series changes. By tracking the temporal process of the occurrence and development of patches in the canopy of the pine forest, the two components of the time dimension and the space dimension are extracted to construct a detector, thereby achieving high-precision early diagnosis of pine wood nematode disease, with a low false alarm rate, high universality, and the ability to effectively reduce ecological and economic losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 This is a schematic flow chart of the method for early determination of nematode disease in pine wood provided by the present invention;

[0042] Figure 2 It is a broken line diagram of the patch evaluation parameters of pine trees provided by the present invention;

[0043] Figure 3 This is a schematic structural diagram of the device for early determination of nematode disease in pine wood provided by the present invention;

[0044] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0046] The current main method for monitoring pine wilt disease is to collect drone or satellite imagery from a single period, and then use supervised classification algorithms, such as deep semantic segmentation networks, target detection networks, traditional machine learning classification models such as random forest algorithms and support vector machines, to identify and locate discolored wood. The supervised learning method generally involves first selecting training samples to be processed in the area of interest, such as samples of diseased discolored wood and normal trees and other landforms. The selected supervised classification model is then trained to obtain a usable classifier, which is then used to classify and identify images of the monitored area. In addition, early diagnosis of pine wilt disease can be carried out by collecting hyperspectral images, screening sensitive bands, and constructing specific indices.

[0047] However, the discolored wood detected by the above monitoring method is often in the late stages of the disease, making it difficult to diagnose the disease at an early stage. Moreover, the infected wood detected by this method does not necessarily die from infection with pine wilt disease. The color of the first-phase image is difficult to distinguish from the early stages of infection, because the early stages of infection generally do not have very significant color changes. In the middle and late stages of infection, the entire pine tree dies and the leaves turn yellow, which is significantly different from normal pine wood.

[0048] Due to the lack of effective prevention and control measures, infected pine trees often die within four weeks, causing widespread forest destruction within two to three years, resulting in significant environmental, ecological, and economic losses. Currently, the primary means of pine wood prevention and control lies in early detection and removal, as well as control of infected areas to prevent the rapid spread of nematode disease. For this purpose, timely monitoring of infected trees is crucial.

[0049] In addition, stress from the natural environment and other pests and diseases can also cause the death of pine trees, making them easily misjudged.

[0050] Hyperspectral imaging methods also have the problem of difficulty in distinguishing whether the pine wood is infected with pine wilt disease, which ultimately leads to the death of pine wood. At present, there is no method to use time-series visible light data to carry out early diagnosis of pine wilt disease.

[0051] The following combination Figure 1-Figure 4 The present invention provides a method and device for early determination of nematode disease in pine wood.

[0052] Figure 1 Schematic diagram of the process of the method for early determination of nematode disease in pine provided by the present invention, as shown in FIG. Figure 1 As shown, including but not limited to the following steps:

[0053] First, in step S1, monitoring images of the target pine forest at multiple observation times are acquired.

[0054] A drone was used to collect high-resolution visible light images of the target pine forest canopy from a bird's-eye view at n observation times within a month, with the interval between each two adjacent observation times being greater than 4 days.

[0055] The visible light image can be a whole canopy map of the target pine forest or multiple canopy maps of each partition. After position calibration and stitching and cropping, the whole canopy map of the target pine forest can be obtained.

[0056] The entire canopy image can be used as a monitoring image; or the monitoring image can be obtained after noise reduction and image enhancement are performed on the entire canopy image.

[0057] By collecting high-resolution canopy images of pine forests using drones and identifying lesions based on this, we can obtain the entire process of pine wilt disease occurring in the canopy of a single pine tree. This can avoid the problem of using low-resolution image data to perceive small lesions, which leads to the identification of infected trees in the middle and late stages of infection, thereby improving the accuracy of early diagnosis.

[0058] Typically, the color change rate of a pine tree's canopy changes significantly when the canopy center and the canopy edge are infected. Disease develops more quickly in the center, and the entire tree dies quickly. Disease spreads more slowly in the canopy edge. Based on this characteristic, real-time monitoring and feedback on the disease status of pine trees is provided by calculating the distance between the lesion and the canopy center.

[0059] Since the pathogenesis of pine wilt disease is a rapidly accelerating process, that is, compared with environmental stress and other diseases, the spread of lesions after pine wilt disease infection is accelerated. Therefore, by calculating the diffusion rate of lesions in time series, early diagnosis of pine wilt disease infection can be achieved.

[0060] Furthermore, in step S2, feature point extraction is performed on each monitoring image to obtain points of the same name in each monitoring image.

[0061] Among them, after extracting the feature points of each monitoring image, each monitoring image is matched to obtain the same-name points of each monitoring image. The more the number of same-name points, the higher the corresponding pixel correction accuracy. Since the early identification of pine nematode disease has high requirements for accuracy, at least 7 same-name points are determined for each monitoring image.

[0062] Optionally, extracting feature points from each monitoring image to obtain points of the same name in each monitoring image includes:

[0063] Extract feature points from each monitoring image to determine the feature points of each monitoring image;

[0064] The feature points of the monitoring images collected at the initial observation time among the multiple observation times are used to match the feature points of each monitoring image to obtain the points with the same name in each monitoring image.

[0065] For example, in OpenCV, the scale-invariant feature transform (SIFT) / random sample consensus (RANSC) algorithm is used to extract feature points from each monitoring image. The Fast Library for Approximate Nearest Neighbors (FLANN) algorithm is used to match the feature points of the monitoring image corresponding to the first observation moment with those of the monitoring images corresponding to subsequent observation moments. This allows for overlapping patches on each monitoring image to obtain the same-name points between the monitoring image corresponding to the first observation moment and the monitoring images corresponding to the subsequent observation moments. Lowe recommends a ratio threshold of 0.6 for FLANN matching.

[0066] According to the method for early determination of pine nematode disease provided by the present invention, feature points of monitoring images are extracted and then matched to obtain the same-name points of each monitoring image, providing a basis for pixel coordinate calibration of the monitoring images.

[0067] Furthermore, in step S3, coordinate correction is performed on all monitoring images based on the same-name points to determine reference information of each pine tree and patch parameters of each patch.

[0068] According to the same-name points of each monitoring image, the monitoring image corresponding to the first observation moment is used as the benchmark to perform coordinate correction on the monitoring images corresponding to other observation moments, and the canopy of each pine tree and the patches on the canopy of each pine tree are delineated on each corrected image.

[0069] Optionally, the coordinate correction is performed on all monitoring images based on the same-name points to determine the reference information of each pine tree and the patch parameters of each patch, including:

[0070] Determine each preset equation using the same-name points of the monitoring image collected at the initial observation time;

[0071] Based on each of the preset equations and all the points of the same name, pixel matching is performed on each monitoring image with the monitoring image collected at the initial observation time to generate a corrected image corresponding to each monitoring image;

[0072] Determine the reference information of each pine tree in each calibrated image, as well as the patch parameters of each patch.

[0073] Among them, based on the quadratic equation y=ax 2 +bx+c, by using the position coordinates of the same-name point in the monitoring image corresponding to the first observation time (x t ,y t ) solves for the parameters a, b, and c to obtain the preset equation for each monitored image. Furthermore, higher-order equations can be used to solve for the parameters of the same-name points in each image.

[0074] For example, for monitoring image 1 corresponding to the first observation moment and monitoring image 2 corresponding to the second observation moment, for a same-name point A shared by monitoring image 1 and monitoring image 2, the coordinates on monitoring image 1 are (x1, y1), and the coordinates on monitoring image 2 are (x2, y2). For any same-name point A on monitoring image 1 and the same-name point B corresponding to the same-name point A on monitoring image 2, since the width w of each monitoring image is the same, the equations can be constructed: A = x1 + y1 × w, B = x2 + y2 × w. Through multiple same-name points on the two monitoring images, multiple groups (A, B) can be obtained. Each group (A, B) is substituted into the quadratic equation y = ax 2 +bx+c, and fitting the coefficients a, b, and c of the quadratic equation, we can get the preset equation of monitoring image 2. Among them, A and B are the results of the one-dimensionalization of the plane coordinates of the same-name points.

[0075] Using the obtained preset equations and all the points of the same name, image matching is performed on the monitoring images corresponding to the subsequent n-1 observation moments in turn. All monitoring images are transformed into the same coordinate system, and an image that matches the monitoring image corresponding to the first observation moment pixel by pixel can be obtained, thereby realizing the temporal registration of the monitoring images.

[0076] Open the image in the image processing software and complete the calibration of each corrected image manually or automatically through training a deep learning model. Specifically: fit the geometric center point of each pine canopy c-j And mark, j = 1, 2, 3, ..., n, find the longest tangent line through the geometric center point, record it as the target tangent Canopy dia-j .

[0077] At the same time, mark the insect-infested area in the canopy of the corrected image corresponding to the j-th observation moment as a patch, and use dia i，j (i=1, 2, 3, ..., p) identifier, i represents the number of the damaged plaque in the correction image corresponding to the jth observation moment, p is the total number of damaged plaques in the correction image, and each dia i，j There are two attributes, one is the distance from the geometric center point dia_can i，j , expressed in pixels; the other is the area of the patch dia_sizei，j , expressed in pixels.

[0078] According to the method for early determination of pine nematode disease provided by the present invention, by performing coordinate correction on each monitoring image, the accurate position of each pine tree and the change of the patch over time can be obtained.

[0079] Furthermore, in step S4, evaluation parameters of the patches on each pine tree are generated according to the reference information of each pine tree and the patch parameters of each patch.

[0080] The reference information may be the position coordinates of a fixed point, or the tangent length data of the pine canopy, or the area of the pine canopy.

[0081] The plaque parameter can be the area of the plaque or the position of the plaque edge boundary.

[0082] The generated evaluation parameters can be the number of patches on each pine tree, the area ratio, and the rate of change of the area or the position of the patch edge boundary.

[0083] Optionally, the reference information includes: a geometric center point and a target tangent line of a pine tree canopy; the patch parameters include: a patch position and a patch area;

[0084] Generating evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch includes:

[0085] Determine, based on the geometric center point of each pine tree canopy in each corrected image and the patch position of each patch in each corrected image, the distance information between each patch and the geometric center point of the pine tree canopy where the patch is located; and determine, based on the patch area of each patch on each pine tree in the corrected image corresponding to two adjacent observation moments and the time difference between the two adjacent observation moments, the patch area parameter;

[0086] Obtaining the relative distance of each patch in the crown of each pine tree according to the distance information of each patch on the crown of each pine tree and the target tangent;

[0087] The patch evaluation parameter of each pine tree in each corrected image is determined according to the patch area parameter and the relative distance of each patch.

[0088] In the canopy of the corrected image corresponding to the jth observation moment, the i-th patch dia i，j Distance to the geometric center point of the pine canopy c-j The distance dia_can i，j , the specific calculation is as follows:

[0089]

[0090] Among them, x Canopyc-j with y Canopyc-j Canopy c-j Coordinates in the x-axis and y-axis directions; x0 and y0 are the dia of the i-th patch respectively. i，j Coordinates in the x- and y-axis directions.

[0091] Calculate the Canopy between patch i and the geometric center point in the corrected image corresponding to the jth observation moment c-j The relative distance RatioDia i，j , the specific calculation is as follows:

[0092] RatioDia i，j =dia_can i，j / Canopy dia -j;

[0093] For each patch in the corrected image corresponding to each observation moment, the patch area dia_size i，j Calculate the patch area dia_size of the patch in the corrected image corresponding to the last observation time j-1 i，j-1 The difference between the two observation times is taken as the ratio of the difference to the time difference (unit: day) between the two observation times. If no patch was found in the last observation corresponding to a patch, the area of the patch obtained in the last observation is considered to be 0.

[0094] For each patch, this value is recorded as RatioSize i，j .

[0095] For any patch i in j observations, calculate RatioDia i，j with RatioSize i，j The product of Deter i，j , as a comprehensive evaluation index that comprehensively considers the size of patch i and its distance from the canopy, is calculated as follows:

[0096] Deter i，j =RatioDia i，j ×RatioSize i，j ;

[0097] Here we construct the weight factor RatioSize of the distance from the canopy center i，j It is aimed at the rapid spread of pine wood nematode fungi by vertical transpiration, and the closer the patch is to the center of the canopy, the faster the spread of pine wood nematode disease.

[0098] For each corrected image corresponding to an observation moment, Deter is obtained using the following formula: i，j The mean of Deter j , as the patch evaluation parameter Deter of the corrected image corresponding to the j-th observation moment j , the specific calculation is as follows:

[0099]

[0100] Thus, the plaque evaluation parameters {Deter1, Deter2, ..., Deter n}.

[0101] According to the method for early determination of pine nematode disease provided by the present invention, by calculating the unified standard evaluation parameters of the patches on the pine tree canopy, the early diagnosis of nematode disease can be achieved more simply and accurately.

[0102] Furthermore, in step S5, the early stage of nematode disease in the target pine forest is determined based on the evaluation parameters.

[0103] According to the generated evaluation parameters, the time period corresponding to the evaluation parameters closest to the early stage of nematode disease in pine trees is taken as the early stage of nematode disease in pine trees in the target pine forest.

[0104] Optionally, determining the early stage of nematode disease in pine trees in the target pine forest based on the evaluation parameters includes:

[0105] Determining the patch change degree of the target pine forest at each two adjacent observation moments according to the patch evaluation parameters on all the corrected images;

[0106] According to the degree of all patch changes, the early stage of nematode infection in the target pine forest is determined.

[0107] The plaque evaluation parameter Deter corresponding to each two adjacent observation moments j With Deter j+1 Calculate the slope k j , j=1, 2, 3,..., n-1.

[0108] If k j If it is greater than 0.5, the target pine forest at the jth observation moment is considered to be infected with pine wilt disease; otherwise, the target pine forest at the jth observation moment is considered not to be infected with pine wilt disease.

[0109] For example, m observation moments of pine wilt infection are selected according to the above method, where m≤n.

[0110] Optionally, determining the early infection time of nematode disease in the target pine forest based on the degree of change of all patches includes:

[0111] When the degree of patch change at any moment in the target pine forest is greater than a preset threshold, determining a time period corresponding to the degree of change in the presence of diseased trees infected with nematodes in the target pine forest;

[0112] In the time periods corresponding to the degree of change in the presence of diseased trees infected with nematodes in the target pine forest, the first time period is determined to be the early stage of nematode infection in the target pine forest.

[0113] For m observation moments of pine trees confirmed to be infected by pine wilt disease, calculate the evaluation parameters of two adjacent patches in time series: l With Deter l+1 The slope is K l , where l = 1, 2, 3, …, m-1.

[0114] Thus, we can get m-1 slopes K1, K2, ..., K m-1 , among the m-1 slopes, the time period between the observation moments corresponding to the first slope K1 greater than 0.5 is determined as the early infection period.

[0115] The development of pine wilt disease varies across both time and space. Lesion size increases significantly after infection, and the closer the infection point is to the canopy center, the faster the disease progresses. To this end, this paper characterizes the disease's development by calculating the change in lesion area over time and the potential rate of disease progression in a target pine forest by calculating the distance between the lesion and the canopy center. Figure 2 is a broken line diagram of the patch evaluation parameters of pine trees provided by the present invention, such as Figure 2 As shown in the figure, as the observation progresses, the severity of the disease will increase, and the Deter n The value increased significantly, while other reasons caused the Deter n The value does not increase significantly, so it can be determined by n The slope can be used to determine whether the infection is pine wood nematode and whether it is in the early stage.

[0116] The method for early determination of pine wood nematode disease provided by the present invention is based on high-resolution visible light data of infected trees with time-series changes. By tracking the temporal process of the occurrence and development of patches in the canopy of the pine forest, the two components of the time dimension and the space dimension are extracted to construct a detector, thereby achieving early diagnosis of pine wood nematode disease with high precision, a low false alarm rate, high universality, and the ability to effectively reduce ecological and economic losses.

[0117] The following describes the device for early determination of pine wood nematode disease provided by the present invention. The device for early determination of pine wood nematode disease described below and the method for early determination of pine wood nematode disease described above can be referenced to each other.

[0118] Figure 3 This is a schematic diagram of the structure of the device for early determination of nematode disease in pine wood provided by the present invention. Figure 3 As shown, including:

[0119] An acquisition module 301 is used to acquire monitoring images of the target pine forest at multiple observation times;

[0120] A feature point extraction module 302 is used to extract feature points from each monitoring image to obtain points with the same name in each monitoring image;

[0121] A coordinate correction module 303 is used to perform coordinate correction on all monitoring images based on the same-name points to determine the reference information of each pine tree and the patch parameters of each patch;

[0122] A generating module 304 is configured to generate evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch;

[0123] The determination module 305 is configured to determine the early stage of nematode disease in the target pine forest according to the evaluation parameters.

[0124] During the operation of the device, the acquisition module 301 acquires monitoring images of the target pine forest at multiple observation times; the feature point extraction module 302 extracts feature points from each monitoring image to obtain the same-name points of each monitoring image; the coordinate correction module 303 performs coordinate correction on all monitoring images based on the same-name points to determine the reference information of each pine tree and the patch parameters of each patch; the generation module 304 generates evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch; the determination module 305 determines the early stage of nematode disease in the target pine forest based on the evaluation parameters.

[0125] The device for early determination of pine wood nematode disease provided by the present invention is based on high-resolution visible light data of infected trees with time-series changes. By tracking the temporal process of the occurrence and development of patches in the canopy of the pine forest, the two components of the time dimension and the space dimension are extracted to construct a detector. The device can achieve early diagnosis of pine wood nematode disease with high precision, low false alarm rate, high universality, and can effectively reduce ecological and economic losses.

[0126] Figure 4 Schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 may call logic instructions in the memory 430 to execute a method for determining early-stage nematode disease in pine trees, the method comprising: acquiring monitoring images of a target pine forest at multiple observation times; extracting feature points from each monitoring image to obtain homonymous points for each monitoring image; performing coordinate correction on all monitoring images based on the homonymous points to determine reference information for each pine tree and patch parameters for each patch; generating evaluation parameters for patches on each pine tree based on the reference information for each pine tree and the patch parameters for each patch; and determining the early-stage nematode disease in the target pine forest based on the evaluation parameters.

[0127] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0128] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the early determination method of pine nematode disease provided by the above methods, the method including: obtaining monitoring images of the target pine forest at multiple observation times; extracting feature points of each monitoring image to obtain the same-name points of each monitoring image; based on the same-name points, performing coordinate correction on all monitoring images to determine the reference information of each pine tree and the patch parameters of each patch; generating evaluation parameters for the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch; and determining the early stage of nematode disease in the target pine forest based on the evaluation parameters.

[0129] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the method for early determination of nematode disease in pine wood provided by the above-mentioned methods, the method comprising: obtaining monitoring images of the target pine forest at multiple observation times; extracting feature points from each monitoring image to obtain the same-name points of each monitoring image; based on the same-name points, performing coordinate correction on all monitoring images to determine the reference information of each pine tree and the patch parameters of each patch; generating evaluation parameters for the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch; and determining the early stage of nematode disease in the target pine forest based on the evaluation parameters.

[0130] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0131] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for early determination of nematode disease in pine wood, characterized in that: include: Acquire monitoring images of the target pine forest at multiple observation times; Extracting feature points from each monitoring image to obtain points with the same name in each monitoring image; Based on the same-name points, coordinate correction is performed on all monitoring images to determine the reference information of each pine tree and the patch parameters of each patch; Based on the same-name points, coordinate correction is performed on all monitoring images to determine the reference information of each pine tree and the patch parameters of each patch, including: Determine each preset equation using the same-name points of the monitoring image collected at the initial observation time among the multiple observation times; Based on each of the preset equations and all the points of the same name, pixel matching is performed on each monitoring image with the monitoring image collected at the initial observation time to generate a corrected image corresponding to each monitoring image; Determine the reference information of each pine tree in each calibrated image, and the patch parameters of each patch; generating evaluation parameters of the patches on each pine tree according to the reference information of each pine tree and the patch parameters of each patch; determining the early stage of nematode disease in the target pine forest based on the evaluation parameters; The reference information includes: the geometric center point and target tangent of the pine tree canopy; the patch parameters include: patch position and patch area; Generating evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch includes: Determine, based on the geometric center point of each pine tree canopy in each corrected image and the patch position of each patch in each corrected image, the distance information between each patch and the geometric center point of the pine tree canopy where the patch is located; and determine, based on the patch area of each patch on each pine tree in the corrected image corresponding to two adjacent observation moments and the time difference between the two adjacent observation moments, the patch area parameter; Obtaining the relative distance of each patch in the crown of each pine tree according to the distance information of each patch on the crown of each pine tree and the target tangent; The patch evaluation parameter of each pine tree in each corrected image is determined according to the patch area parameter and the relative distance of each patch.

2. The method for early determination of nematode disease in pine wood according to claim 1, characterized in that: The extracting feature points from each monitoring image to obtain the same-name points of each monitoring image includes: Extract feature points from each monitoring image to determine the feature points of each monitoring image; The feature points of the monitoring images collected at the initial observation time among the multiple observation times are used to match the feature points of each monitoring image to obtain the points with the same name in each monitoring image.

3. The method for early determination of nematode disease in pine wood according to claim 1, characterized in that: Determining the early stage of nematode disease in pine trees in the target pine forest based on the evaluation parameters includes: Determining the patch change degree of the target pine forest at each two adjacent observation moments according to the patch evaluation parameters on all the corrected images; According to the degree of all patch changes, the early stage of nematode infection in the target pine forest is determined.

4. The method for early determination of nematode disease in pine wood according to claim 3, characterized in that: Determining the early stage of nematode disease in the target pine forest based on the degree of all patch changes includes: When the degree of patch change at any moment in the target pine forest is greater than a preset threshold, determining a time period corresponding to the degree of change in the presence of diseased trees infected with nematodes in the target pine forest; In the time periods corresponding to the degree of change in the presence of diseased trees infected with nematodes in the target pine forest, the first time period is determined to be the early stage of nematode infection in the target pine forest.

5. An early detection device for nematode disease in pine wood, characterized in that: include: An acquisition module is used to obtain monitoring images of the target pine forest at multiple observation times; A feature point extraction module is used to extract feature points from each monitoring image to obtain points with the same name in each monitoring image; A coordinate correction module, configured to perform coordinate correction on all monitoring images based on the same-name points to determine reference information of each pine tree and patch parameters of each patch; Based on the same-name points, coordinate correction is performed on all monitoring images to determine the reference information of each pine tree and the patch parameters of each patch, including: Determine each preset equation using the same-name points of the monitoring image collected at the initial observation time among the multiple observation times; Based on each of the preset equations and all the points of the same name, pixel matching is performed on each monitoring image with the monitoring image collected at the initial observation time to generate a corrected image corresponding to each monitoring image; Determine the reference information of each pine tree in each calibrated image, and the patch parameters of each patch; a generating module, configured to generate evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch; a determination module, configured to determine the early stage of nematode disease in the target pine forest based on the evaluation parameters; The reference information includes: the geometric center point and target tangent of the pine tree canopy; the patch parameters include: patch position and patch area; Generating evaluation parameters of the patches on each pine tree based on the reference information of each pine tree and the patch parameters of each patch includes: Determine, based on the geometric center point of each pine tree canopy in each corrected image and the patch position of each patch in each corrected image, the distance information between each patch and the geometric center point of the pine tree canopy where the patch is located; and determine, based on the patch area of each patch on each pine tree in the corrected image corresponding to two adjacent observation moments and the time difference between the two adjacent observation moments, the patch area parameter; Obtaining the relative distance of each patch in the crown of each pine tree according to the distance information of each patch on the crown of each pine tree and the target tangent; The patch evaluation parameter of each pine tree in each corrected image is determined according to the patch area parameter and the relative distance of each patch.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for early determination of nematode disease in pine wood as described in any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for early determination of nematode disease in pine wood as claimed in any one of claims 1 to 4 is implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for early determination of nematode disease in pine wood as claimed in any one of claims 1 to 4 is implemented.

Images