A pine wood nematode pest situation early warning monitoring system and method

Abstract

The application discloses a pine wood nematode pest situation early warning monitoring system and method, and relates to the technical field of pest control. The system comprises a data acquisition module, an image acquisition module, an activity recognition module, a feeding recognition module, a model establishment module, a resin recognition module, a pest damage recognition module, a pine wood nematode recognition module and an early warning module. The pest damage recognition module selects at least one nematode affected area from a resin existing area by using a resin estimation model and resin volume in the resin existing area. The pine wood nematode recognition module identifies at least one pest situation area where the pine wood nematode exists in the feeding area. The early warning module is provided with the activity recognition module, the feeding recognition module, the model establishment module, the resin recognition module, the pest damage recognition module and the pine wood nematode recognition module, so that the problem that the initial pest situation of the pine wood nematode is very concealed can be solved, the initial pest situation of the pine wood nematode is made explicit, and accurate monitoring can be performed according to the analysis result.

Description

Technical Field

[0001] This invention relates to the field of pest control technology, specifically to a pine wilt nematode infestation early warning and monitoring system and method. Background Technology

[0002] Pine wilt disease, also known as pine blight or pine rot, is a devastating disease caused by the pine wood nematode, which parasitizes pine trees. The nematode is extremely small, invisible to the naked eye, and resembles a miniature roundworm, measuring 0.014-0.016 mm in length. Pine wood nematodes do not spread directly but primarily parasitize the pine sawyer beetle, entering the branches of healthy pine trees through wounds inflicted by the beetle on pine branches.

[0003] Prevention of pine wilt disease is usually achieved by timely detection and repair of early parasitism. However, the early infestation of pine wilt disease is very insidious, with no obvious wounds, making it difficult to detect and posing a great challenge to early warning. Summary of the Invention

[0004] To address the aforementioned technical problems, a pine wilt nematode infestation early warning and monitoring system and method are provided. This technical solution solves the problems mentioned in the background section.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A pine wilt nematode infestation early warning and monitoring system includes:

[0007] The data acquisition module acquires at least one pine tree in the monitoring area and at least one vector of pine wilt nematode in the monitoring area, wherein the vector is a different species of longhorn beetle in the monitoring area.

[0008] An image acquisition module acquires at least one sample image of the erosion marks on the propagation medium;

[0009] An activity recognition module identifies the feeding areas of the propagating medium on the surface of the pine tree to obtain key areas;

[0010] A gnawing recognition module, which identifies at least one gnawing area of ​​the propagating medium in a key area based on a sample image;

[0011] The model building module pre-forms a resin identification model and a resin estimation model.

[0012] A resin identification module, which uses a resin identification model to identify at least one area where resin is present and the volume of resin in the area on the surface of a pine tree;

[0013] The pest identification module uses a resin estimation model and the resin volume in the resin presence area to select at least one nematode-affected area from the resin presence area.

[0014] A pine wood nematode identification module, which identifies at least one infestation area where pine wood nematodes exist in the feeding area based on the distribution of the nematode-affected area;

[0015] The early warning module does not issue an early warning when the number of insect-infested areas is 0; otherwise, it issues an early warning for insect-infested areas in the pine trees.

[0016] Preferably, obtaining at least one sample image of the grazing marks on the propagation medium includes the following steps:

[0017] Obtain at least one historical image of the pine tree being eaten by the medium, and identify historical wounds on the pine tree in the historical image;

[0018] Extract the grazing marks along the contour line of the historical wound, delete the part of the contour line except for the grazing marks, and keep only the grazing marks to obtain a sample image of the grazing marks.

[0019] Preferably, identifying the feeding area of ​​the proliferating medium on the surface of the pine tree to obtain the key area includes the following steps:

[0020] Beforehand, obtain images of the excrement of the transmission medium, and extract the range of pixel values ​​of the pixels in the excrement from the excrement images as the sample range;

[0021] In the actual image of the pine tree surface, pixels whose pixel values ​​belong to the sample range are taken as target points, and adjacent target points are aggregated to form at least one discharge point.

[0022] The average pixel value of each excretion point is taken to obtain the feature value of the excretion point. The excretion points are connected in order of increasing feature value to obtain the activity path of the transmission medium.

[0023] The maximum feeding value between adjacent excretion intervals of the transmission medium is obtained in advance as the feeding limit;

[0024] The minimum feeding speed of the transmission medium is obtained, and the upper limit of feeding speed is divided by the minimum feeding speed to obtain the upper limit of feeding time. The average speed of the transmission medium during feeding is obtained as the characteristic speed.

[0025] The upper limit of feeding time is multiplied by the characteristic speed to obtain the limit distance. At least one sampling point is uniformly selected on the activity path. The area where the distance to the sampling point is less than the limit distance is used as the sampling area of ​​the sampling point. The sampling areas are merged to obtain the key area.

[0026] Preferably, identifying at least one biting area of ​​the propagating medium in a key area based on the sample image includes the following steps:

[0027] At least one normal texture image of the pine tree surface is obtained in advance;

[0028] Take at least one reference point uniformly in the (0, 100) interval, and scale the normal texture image according to the ratio of the reference point to obtain the scaled texture image.

[0029] The difference between adjacent pixels in a normal texture image is calculated and the absolute value is taken to obtain at least one sampled value. The maximum value of the at least one sampled value is used as the threshold value.

[0030] Calculate the absolute value of the difference between the pixel values ​​of adjacent pixels in the key area to obtain the actual difference. Take any one of the adjacent pixels in the key area whose actual difference is greater than the threshold value as the sampling point.

[0031] Adjacent collection points are aggregated to form at least one collection profile;

[0032] Delete the acquisition contours that are consistent with the scaled texture image, and use the area enclosed by the deleted acquisition contours as the suspected area.

[0033] The sample image of the grazing marks is scaled up according to the ratio of the reference point to obtain the scaled sample image;

[0034] If the grazing marks in the scaled-up sample image appear at the location traversed by the outline of the suspected region, then the suspected region is considered as the grazing region; otherwise, no processing is performed.

[0035] Preferably, the pre-formed resin recognition model includes the following steps:

[0036] Obtain the range of resin thickness values, divide the resin thickness range into equal intervals, and obtain at least one data point;

[0037] Obtain the resin with a thickness equal to the value of the data point as the sampled resin, and obtain the average value of the pixel values ​​of all pixels in the sampled resin image as the reference value of the sampled resin;

[0038] The baseline value of the sampled pine resin is paired with the data points and fitted to obtain the identification fitting function, where the baseline value is the independent variable and the data points are the dependent variable.

[0039] Preferably, the pre-formed resin estimation model includes the following steps:

[0040] Screening is performed on at least one pest that causes pest damage to pine trees to obtain at least one target pest, wherein the target pest is a pest that causes pest damage on the surface of pine trees.

[0041] Obtain the area of ​​damage caused by the target pest on the surface of the pine tree, and divide the area of ​​damage evenly to obtain at least one dividing point.

[0042] Under the condition that the damage area of ​​the target pest is equal to the value at the dividing point, the volume of resin produced at the damage area is obtained as the reference volume.

[0043] The values ​​at the cut-off points are paired with the baseline volume and fitted to obtain the estimated fitting function, where the values ​​at the cut-off points are the independent variables and the baseline volume is the dependent variable.

[0044] The estimated fitting functions corresponding to all target pests are summarized into a resin estimation model.

[0045] Preferably, the step of using a resin identification model to identify at least one resin-containing area and the resin volume within that area on the pine tree surface includes the following steps:

[0046] The range of pixel values ​​for each pixel in the image is obtained in advance and used as the feature range.

[0047] In the actual image of a pine tree, pixels whose pixel values ​​belong to the feature range are taken as feature pixels, and adjacent feature pixels are aggregated to form at least one area where resin exists.

[0048] The area where pine resin exists is uniformly divided to obtain at least one local region block. The average value of the pixel values ​​of the pixels in the local region block is taken to obtain the local value.

[0049] Substituting the local values ​​into the identification fitting function yields the local thickness. Multiplying the local thickness by the base area of ​​the local region block yields the local volume. The local volumes of the local region blocks in the resin-containing region are then summed to obtain the resin volume in the resin-containing region.

[0050] Preferably, selecting at least one nematode-affected area from the resin-containing area includes the following steps:

[0051] Identify areas where resin with surface damage exists as target areas;

[0052] Substitute the surface damage area of ​​the target region into the estimation fitting function to obtain the target volume;

[0053] When the volume of each target is different from the volume of pine resin in the target area, the target area is considered as the nematode-affected area.

[0054] Preferably, identifying at least one infestation area where pine wood nematodes exist within the feeding area based on the distribution of the nematode-affected area includes the following steps:

[0055] Areas containing pine resin will be considered as potential gnaw areas.

[0056] Based on historical impact data of pine wood nematode, the maximum impact distance of pine wood nematode was obtained;

[0057] The alternative feeding areas that are no more than the maximum impact distance from the nematode-affected area are summarized to form a feeding area set of the nematode-affected area.

[0058] The intersection of the sets of feeding regions of any two nematode-affected areas is obtained as the feeding intersection.

[0059] All potential feeding areas where feeding overlap are designated as insect infestation areas.

[0060] A method for early warning and monitoring of pine wilt nematode infestation, used to implement the aforementioned pine wilt nematode infestation early warning and monitoring system, comprising:

[0061] Obtain at least one pine tree within the monitoring area, and obtain at least one vector of pine wood nematode within the monitoring area, wherein the vector is a different species of longhorn beetle within the monitoring area;

[0062] Obtain at least one sample image of the erosion marks on the transmission medium;

[0063] Identify the feeding areas of the vectors on the surface of pine trees to obtain key areas;

[0064] Based on sample images, at least one erosion area of ​​the propagating medium was identified in the key areas;

[0065] Pre-form a resin identification model and a resin estimation model;

[0066] Using a resin identification model, at least one area where resin is present and the volume of resin in that area can be identified on the surface of a pine tree.

[0067] Using a resin estimation model and the resin volume in the resin-containing area, at least one nematode-affected area is selected from the resin-containing area.

[0068] Based on the distribution of the nematode-affected areas, at least one infestation area where pine wood nematodes are present can be identified in the feeding area.

[0069] When the number of insect-infested areas is 0, no warning is issued; otherwise, a warning is issued for insect-infested areas in the pine trees.

[0070] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0071] By setting up activity identification modules, feeding identification modules, model building modules, resin identification modules, pest identification modules, and pine wilt nematode identification modules, the volume of pine resin in areas where pine resin exists is identified. In key areas, at least one feeding area where the vector feeds is identified. Thus, by observing the impact of pine wilt nematodes on pine resin, areas where pine resin exists are screened to obtain nematode-affected areas. Based on the distribution of nematode-affected areas, infestation areas are identified, and targeted early warnings are issued. This solves the problem of the very concealed infestation of pine wilt nematodes in the early stages, making the analysis of the infestation of pine wilt nematodes in the early stages explicit, and enabling accurate monitoring based on the analysis results. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of the pine wilt nematode infestation early warning and monitoring system of the present invention;

[0073] Figure 2 This is a schematic diagram of the process for obtaining at least one sample image of the grazing marks on the propagation medium according to the present invention;

[0074] Figure 3 This is a schematic diagram illustrating the process of identifying the feeding area of ​​the propagating medium on the surface of a pine tree to obtain the key area according to the present invention.

[0075] Figure 4 This is a schematic diagram of the process of identifying at least one erosion area of ​​the propagating medium in a key area based on a sample image according to the present invention.

[0076] Figure 5 This is a schematic diagram of the process for pre-forming a resin identification model according to the present invention;

[0077] Figure 6 This is a schematic diagram of the process for pre-forming a resin estimation model according to the present invention;

[0078] Figure 7 This is a schematic diagram illustrating the process of using a resin identification model to identify at least one resin-containing area and the volume of resin in the resin-containing area on the surface of a pine tree according to the present invention.

[0079] Figure 8 This is a schematic diagram of the process for selecting at least one nematode-affected area from the area where pine resin is present according to the present invention;

[0080] Figure 9 This is a schematic diagram of the process of identifying at least one infestation area of ​​pine wood nematode in the feeding area based on the distribution of the nematode-affected area according to the present invention. Detailed Implementation

[0081] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0082] Reference Figure 1 As shown, a pine wilt nematode infestation early warning and monitoring system includes:

[0083] The data acquisition module acquires at least one pine tree in the monitoring area and at least one vector of pine wilt nematode in the monitoring area, wherein the vector is a different species of longhorn beetle in the monitoring area.

[0084] An image acquisition module acquires at least one sample image of the erosion marks on the propagation medium;

[0085] An activity recognition module identifies the feeding areas of the propagating medium on the surface of the pine tree to obtain key areas;

[0086] A gnawing recognition module, which identifies at least one gnawing area of ​​the propagating medium in a key area based on a sample image;

[0087] The model building module pre-forms a resin identification model and a resin estimation model.

[0088] A resin identification module, which uses a resin identification model to identify at least one area where resin is present and the volume of resin in the area on the surface of a pine tree;

[0089] The pest identification module uses a resin estimation model and the resin volume in the resin presence area to select at least one nematode-affected area from the resin presence area.

[0090] A pine wood nematode identification module, which identifies at least one infestation area where pine wood nematodes exist in the feeding area based on the distribution of the nematode-affected area;

[0091] The early warning module does not issue an early warning when the number of insect-infested areas is 0; otherwise, it issues an early warning for insect-infested areas in the pine trees.

[0092] This approach primarily focuses on detecting pine wood nematodes by understanding their impact on pine resin. After parasitizing pine trees, the nematode reduces resin secretion. The role of resin is to control pests within the pine tree; when pests are present, resin is secreted at that location. However, when affected by pine wood nematodes, resin secretion is insufficient. Since pests are not always observable on the pine tree surface, it's necessary to screen the resin-producing areas for observation. Subsequent steps will address this; otherwise, insufficient estimation of pest activity at resin-producing areas makes it difficult to determine if resin secretion is normal, thus hindering the identification of areas with abnormal secretion. Furthermore, because this is image recognition rather than 3D modeling, obtaining precise resin quantity is challenging, yet this quantity is crucial for analysis and judgment. Subsequent steps will address these issues.

[0093] Reference Figure 2 As shown, obtaining at least one sample image of the grazing marks on the propagation medium includes the following steps:

[0094] Obtain at least one historical image of the pine tree being eaten by the medium, and identify historical wounds on the pine tree in the historical image;

[0095] Extract the grazing marks along the contour line of the historical wound, delete the part of the contour line except for the grazing marks, and keep only the grazing marks to obtain a sample image of the grazing marks.

[0096] Reference Figure 3 As shown, identifying the feeding areas of the vector on the surface of the pine tree to obtain the key areas involves the following steps:

[0097] Beforehand, obtain images of the excrement of the transmission medium, and extract the range of pixel values ​​of the pixels in the excrement from the excrement images as the sample range;

[0098] In the actual image of the pine tree surface, pixels whose pixel values ​​belong to the sample range are taken as target points, and adjacent target points are aggregated to form at least one discharge point.

[0099] The average pixel value of each excretion point is taken to obtain the feature value of the excretion point. The excretion points are connected in order of increasing feature value to obtain the activity path of the transmission medium.

[0100] The maximum feeding value between adjacent excretion intervals of the transmission medium is obtained in advance as the feeding limit;

[0101] The minimum feeding speed of the transmission medium is obtained, and the upper limit of feeding speed is divided by the minimum feeding speed to obtain the upper limit of feeding time. The average speed of the transmission medium during feeding is obtained as the characteristic speed.

[0102] The upper limit of feeding time is multiplied by the characteristic speed to obtain the limit distance. At least one sampling point is uniformly selected on the activity path. The area where the distance to the sampling point is less than the limit distance is used as the sampling area of ​​the sampling point. The sampling areas are merged to obtain the key area.

[0103] The activity area of ​​the media is not the entire pine tree, but its activity area is also difficult to determine. Its excrement can help to obtain the activity area, because the vicinity of the excrement is its activity area. However, the excrement needs to be sorted in chronological order to obtain its excretion path. Since the excrement changes over time, its color is either darker or lighter. In either case, sorting the excrement by pixel value will yield the path of excrement produced over time. The only difference is whether it is in chronological order or in reverse chronological order, but this is not important for this solution. Therefore, once the activity path of the media is determined, its total activity area can be determined based on the activity path, which can then be used as the key area.

[0104] Reference Figure 4 As shown, identifying at least one biting area of ​​the propagating medium in a key region based on sample images includes the following steps:

[0105] At least one normal texture image of the pine tree surface is obtained in advance;

[0106] Take at least one reference point uniformly in the (0, 100) interval, and scale the normal texture image according to the ratio of the reference point to obtain the scaled texture image.

[0107] The difference between adjacent pixels in a normal texture image is calculated and the absolute value is taken to obtain at least one sampled value. The maximum value of the at least one sampled value is used as the threshold value.

[0108] Calculate the absolute value of the difference between the pixel values ​​of adjacent pixels in the key area to obtain the actual difference. Take any one of the adjacent pixels in the key area whose actual difference is greater than the threshold value as the sampling point.

[0109] Adjacent collection points are aggregated to form at least one collection profile;

[0110] Delete the acquisition contours that are consistent with the scaled texture image, and use the area enclosed by the deleted acquisition contours as the suspected area.

[0111] The sample image of the grazing marks is scaled up according to the ratio of the reference point to obtain the scaled sample image;

[0112] If the grazing marks in the scaled-up sample image appear at the location traversed by the outline of the suspected region, then the suspected region is considered as the grazing region; otherwise, no processing is performed.

[0113] When performing image recognition, it is necessary to consider the possibility of inconsistent image sizes. Therefore, the image used for comparison needs to be enlarged. Typically, the scaling ratio is only within the range of (0, 100). By scaling using a reference point generated in the range of (0, 100), there will inevitably be an image in the enlarged image that has the same size ratio as the image of the suspected area. By comparing this image, it is possible to identify whether there are longhorn beetle grazing marks in the suspected area. It should be noted that during the comparison, only the area through which the outline of the suspected area passes needs to be compared. It is not necessary to compare all positions within the suspected area, because when longhorn beetles graze, they mainly graze the edges of the grazing area, thereby expanding the grazing area. The comparison mainly depends on their grazing habits.

[0114] Reference Figure 5 As shown, the pre-formation of a resin recognition model includes the following steps:

[0115] Obtain the range of resin thickness values, divide the resin thickness range into equal intervals, and obtain at least one data point;

[0116] Obtain the resin with a thickness equal to the value of the data point as the sampled resin, and obtain the average value of the pixel values ​​of all pixels in the sampled resin image as the reference value of the sampled resin;

[0117] The baseline value of the sampled pine resin is paired with the data points and fitted to obtain the identification fitting function, where the baseline value is the independent variable and the data points are the dependent variable.

[0118] Because of the varying thickness of pine resin, its color also varies, resulting in different pixel values. Based on this, a corresponding model can be built, and the thickness of pine resin in a planar image can be identified according to the model and pixel values, thereby obtaining the volume of pine resin in the region.

[0119] Reference Figure 6 As shown, the pre-formation of a resin estimation model includes the following steps:

[0120] Screening is performed on at least one pest that causes pest damage to pine trees to obtain at least one target pest, wherein the target pest is a pest that causes pest damage on the surface of pine trees.

[0121] Obtain the area of ​​damage caused by the target pest on the surface of the pine tree, and divide the area of ​​damage evenly to obtain at least one dividing point.

[0122] Under the condition that the damage area of ​​the target pest is equal to the value at the dividing point, the volume of resin produced at the damage area is obtained as the reference volume.

[0123] The values ​​at the cut-off points are paired with the baseline volume and fitted to obtain the estimated fitting function, where the values ​​at the cut-off points are the independent variables and the baseline volume is the dependent variable.

[0124] The estimated fitting functions corresponding to all target pests are summarized into a resin estimation model.

[0125] Different pests have different characteristics; some live below the surface of pine trees, while others live on the surface. Therefore, the damage they cause is visible, and the severity of the pest can be assessed based on the damage. When pine trees are not affected by pine wilt nematodes, their resin secretion is normal and will vary according to the severity of the pest. Therefore, a resin estimation model can be established to estimate the resin secretion of trees unaffected by pine wilt nematodes. This estimate can then be compared with the actual amount of resin to determine the areas where resin is present due to pine wilt nematodes.

[0126] Reference Figure 7 As shown, using a resin identification model, identifying at least one resin-containing area and the resin volume within that area on the surface of a pine tree includes the following steps:

[0127] The range of pixel values ​​for each pixel in the image is obtained in advance and used as the feature range.

[0128] In the actual image of a pine tree, pixels whose pixel values ​​belong to the feature range are taken as feature pixels, and adjacent feature pixels are aggregated to form at least one area where resin exists.

[0129] The area where pine resin exists is uniformly divided to obtain at least one local region block. The average value of the pixel values ​​of the pixels in the local region block is taken to obtain the local value.

[0130] Substituting the local values ​​into the identification fitting function yields the local thickness. Multiplying the local thickness by the base area of ​​the local region block yields the local volume. The local volumes of the local region blocks in the resin-containing region are then summed to obtain the resin volume in the resin-containing region.

[0131] Reference Figure 8 As shown, selecting at least one nematode-affected area from the area where pine resin is present includes the following steps:

[0132] Identify areas where resin with surface damage exists as target areas;

[0133] Substitute the surface damage area of ​​the target region into the estimation fitting function to obtain the target volume;

[0134] When the volume of each target is different from the volume of pine resin in the target area, the target area is considered as the nematode-affected area.

[0135] Since the estimated fitting function corresponds to the target pest, and the impact of each target pest is different, the damage area caused by each pest has a different impact on the pine tree. The difference lies in the depth of the wound. Therefore, the amount of resin secreted by different target pests is different. Thus, it is necessary to form an estimated fitting function for each target pest separately. When comparing, it is necessary to compare the target volume obtained by each estimated fitting function with the resin volume in the target area. Only when both are different can the target area be regarded as the nematode-affected area.

[0136] Reference Figure 9 As shown, based on the distribution of nematode-affected areas, identifying at least one infestation area where pine wood nematodes are present in the feeding area includes the following steps:

[0137] Areas containing pine resin will be considered as potential gnaw areas.

[0138] Based on historical impact data of pine wood nematode, the maximum impact distance of pine wood nematode was obtained;

[0139] The alternative feeding areas that are no more than the maximum impact distance from the nematode-affected area are summarized to form a feeding area set of the nematode-affected area.

[0140] The intersection of the sets of feeding regions of any two nematode-affected areas is obtained as the feeding intersection.

[0141] All potential feeding areas where feeding overlap are designated as insect infestation areas.

[0142] Pine wood nematode cannot spread directly on its own and can only be spread through longhorn beetles. Therefore, it will only appear in the areas where it gnaws. However, pine wood nematodes can only be present in the areas where pine resin is present. Therefore, it is sufficient to analyze the candidate gnaw areas.

[0143] When pine wood nematodes are present in the candidate feeding area, they will not only affect one nematode-affected area, because their influence range is relatively wide. Therefore, they will inevitably appear at the intersection of the feeding areas of the two nematode-affected areas.

[0144] Here, it is important to note that taking the intersection of the sets of feeding regions of any two nematode-affected areas means that, assuming a, b, and c are the sets of feeding regions of all nematode-affected areas, the intersection of a and b, the intersection of b and c, and the intersection of a and c each yield a feeding intersection, for a total of three feeding intersections. The same approach is taken for cases with a larger number of feeding regions.

[0145] A method for early warning and monitoring of pine wilt nematode infestation, used to implement the aforementioned pine wilt nematode infestation early warning and monitoring system, comprising:

[0146] Obtain at least one pine tree within the monitoring area, and obtain at least one vector of pine wood nematode within the monitoring area, wherein the vector is a different species of longhorn beetle within the monitoring area;

[0147] Obtain at least one sample image of the erosion marks on the transmission medium;

[0148] Identify the feeding areas of the vectors on the surface of pine trees to obtain key areas;

[0149] Based on sample images, at least one erosion area of ​​the propagating medium was identified in the key areas;

[0150] Pre-form a resin identification model and a resin estimation model;

[0151] Using a resin identification model, at least one area where resin is present and the volume of resin in that area can be identified on the surface of a pine tree.

[0152] Using a resin estimation model and the resin volume in the resin-containing area, at least one nematode-affected area is selected from the resin-containing area.

[0153] Based on the distribution of the nematode-affected areas, at least one infestation area where pine wood nematodes are present can be identified in the feeding area.

[0154] When the number of insect-infested areas is 0, no warning is issued; otherwise, a warning is issued for insect-infested areas in the pine trees.

[0155] Furthermore, this solution also proposes a storage medium on which a computer-readable program is stored, which, when invoked, executes the aforementioned pine wilt nematode infestation early warning and monitoring system.

[0156] It is understandable that the storage medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid-state drive (SSD).

[0157] In summary, the advantages of this invention are as follows: by setting up an activity identification module, a feeding identification module, a model building module, a resin identification module, a pest identification module, and a pine wilt nematode identification module, the volume of pine resin in the resin-containing area is identified, and at least one feeding area of ​​the vector is identified in key areas. Thus, by observing the impact of pine wilt nematodes on pine resin, the resin-containing area is screened to obtain the nematode-affected area. Based on the distribution of the nematode-affected area, the pest infestation area is obtained, and targeted early warning is provided. This solves the problem of the very concealed nature of the early pine wilt nematode infestation, making the early pine wilt nematode infestation analysis explicit, and accurately monitoring based on the analysis results.

[0158] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A pine wood nematode pest situation early warning monitoring system, characterized by, The method comprises the following steps: a data acquisition module acquires at least one pine tree in a monitoring area, acquires at least one transmission medium of pine wood nematode in the monitoring area, and the transmission medium is a different kind of beetle in the monitoring area; an image acquisition module acquires at least one sample image of a feeding trace of the transmission medium; an activity recognition module recognizes a feeding area of the transmission medium on the surface of the pine tree, and obtains a key area; a feeding recognition module recognizes at least one feeding area of the transmission medium based on the sample image in the key area; a model establishment module preforms a resin recognition model and a resin estimation model; a resin recognition module recognizes at least one resin existing area and resin volume in the resin existing area on the surface of the pine tree by using the resin recognition model; a pest recognition module selects at least one nematode affected area from the resin existing area by using the resin estimation model and the resin volume in the resin existing area; a pine wood nematode recognition module recognizes at least one pest situation area in which the pine wood nematode exists in the feeding area according to the distribution of the nematode affected area; a warning module does not issue a warning when the number of pest situation areas is 0, otherwise, the pest situation areas in the pine tree are warned.

2. The pine wood nematode outbreak warning and monitoring system according to claim 1, wherein The method comprises the following steps: at least one historical image of the transmission medium feeding is acquired, and a historical wound of the pine tree is recognized in the historical image; a feeding trace is extracted along a contour line of the historical wound, a part other than the feeding trace at the contour line is deleted, only the feeding trace is reserved, and a sample image of the feeding trace is obtained.

3. The system according to claim 2, wherein the system further comprises a plurality of sensors for detecting the presence of the pine wood nematode in the soil. The method comprises the following steps: a feces image of the transmission medium is pre-acquired, a pixel value range of pixel points in the feces image is acquired as a sample range; in an actual image of the surface of the pine tree, pixel points with pixel values belonging to the sample range are taken as target points, adjacent target points are aggregated to form at least one excretion point; a mean value of pixel values of pixel points of the excretion point is obtained as a characteristic value of the excretion point, and the excretion points are connected in order from small to large characteristic values to obtain an activity path of the transmission medium; a maximum value of feeding between adjacent excretion intervals of the transmission medium is pre-acquired as an upper limit of feeding; a minimum feeding speed of the transmission medium is acquired, the upper limit of feeding is divided by the minimum feeding speed to obtain an upper limit of feeding duration, and an average speed of the transmission medium feeding is acquired as a characteristic speed; the upper limit of feeding duration is multiplied by the characteristic speed to obtain a limit distance, at least one sampling point is uniformly taken on the activity path, a region with a distance to the sampling point less than the limit distance is taken as a sampling area of the sampling point, and the sampling areas are merged to obtain the key area.

4. The pine wood nematode outbreak early warning monitoring system according to claim 3, wherein The method comprises the following steps: Pre-acquire at least one normal texture image of the pine tree surface; Uniformly take at least one reference point in the interval (0, 100), and scale the normal texture image according to the proportion of the reference point to obtain a scaled texture image; Difference the pixel values of adjacent pixel points in the normal texture image and take the absolute value to obtain at least one sample value, and take the maximum value of the at least one sample value as a critical value; Calculate the absolute value of the difference between the pixel values of adjacent pixel points in the key area to obtain an actual distance, and take any one of the adjacent pixel points in the key area with an actual distance greater than the critical value as a collection point; Aggregate adjacent collection points to form at least one collection contour; Delete the collection contour consistent with the scaled texture image, and take the area surrounded by the deleted collection contour as a suspected area; Scale the sample image of the gnawing trace according to the proportion of the reference point to obtain a sample scaled image; When the gnawing trace in the sample scaled image appears at the position passed by the outline of the suspected area, the suspected area is taken as a gnawing area, otherwise, no treatment is made.

5. The pine wood nematode outbreak early warning monitoring system according to claim 4, wherein The pine tar identification model formed in advance includes the following steps: Obtain the thickness value range of the pine tar, evenly divide the thickness value range of the pine tar, and obtain at least one data point; Obtain the pine tar with a thickness equal to the value of the data point as a sample pine tar, and obtain the average value of the pixel values of all pixel points in the sample pine tar image as the reference value of the sample pine tar; Pair the reference value of the sample pine tar with the data point and fit to obtain an identification fitting function, wherein the reference value is the independent variable and the data point is the dependent variable.

6. The pine wood nematode disease warning monitoring system according to claim 5, wherein The pine tar estimation model formed in advance includes the following steps: Screen at least one target pest from at least one pest that affects the pine tree to produce pests, wherein the target pest is a pest that produces pests on the surface of the pine tree; Obtain the damage area range produced by the target pest on the surface of the pine tree, and evenly divide the damage area range to obtain at least one division point; Under the condition that the damage area of the target pest is equal to the value at the division point, obtain the volume of the pine tar produced at the damage area as a reference volume; Pair the value at the division point with the reference volume and fit to obtain an estimation fitting function, wherein the value at the division point is the independent variable and the reference volume is the dependent variable; Summarize all the estimation fitting functions corresponding to the target pests as a pine tar estimation model.

7. The pine wood nematode outbreak early warning monitoring system according to claim 6, wherein The pine tar identification model is used to identify at least one pine tar existing area and the volume of the pine tar in the pine tar existing area on the surface of the pine tree, which includes the following steps: Pre-acquire the value range of the pixel value of the pixel point of the pine tar in the image as a feature range; In the actual image of the pine tree, take the pixel point with a pixel value belonging to the feature range as a feature pixel point, and aggregate adjacent feature pixel points to form at least one pine tar existing area; Uniformly divide the pine tar existing area to obtain at least one local area block, and take the average value of the pixel values of the pixel points in the local area block to obtain a local value; The local value is substituted into the identification fitting function to obtain a local thickness, the local thickness is multiplied by a bottom area of the local area block to obtain a local volume, and the local volumes of the local area blocks in the turpentine existing area are accumulated to obtain a turpentine volume in the turpentine existing area.

8. The pine wood nematode disease warning monitoring system according to claim 7, wherein The selecting at least one nematode affected area from the turpentine existing area comprises the following steps: The turpentine existing area where the surface damage exists is identified as a target existing area; The surface damage area of the target existing area is substituted into the estimation fitting function to obtain a target volume; When each target volume is different from the turpentine volume in the target existing area, the target existing area is taken as the nematode affected area.

9. The pine wood nematode disease warning and monitoring system according to claim 8, wherein, The at least one pest situation area where the pine wood nematode exists in the feeding area is identified according to the distribution of the nematode affected area, and the at least one pest situation area comprises the following steps: The feeding area where the turpentine exists is taken as a candidate feeding area; The maximum influence distance of the pine wood nematode is obtained based on historical influence data of the pine wood nematode; The candidate feeding areas with a distance to the nematode affected area not exceeding the maximum influence distance are summarized to form a feeding area set of the nematode affected area; The feeding area sets of any two nematode affected areas are intersected to obtain a feeding intersection; The candidate feeding areas in all feeding intersections are taken as the pest situation area.

10. A method for monitoring and warning of the pine wood nematode pest situation, for implementing the pine wood nematode pest situation monitoring and warning system according to any one of claims 1-9, characterized in that, The method comprises the following steps: At least one pine tree in a monitoring area is obtained, at least one transmission medium of the pine wood nematode in the monitoring area is obtained, and the transmission medium is a different kind of beetle in the monitoring area; At least one sample image of a feeding trace of the transmission medium is obtained; An eating area of the transmission medium on the surface of the pine tree is identified to obtain a key area; At least one feeding area where the transmission medium feeds is identified in the key area based on the sample image; A turpentine identification model is formed in advance, and a turpentine estimation model is formed in advance; The turpentine identification model is used to identify at least one turpentine existing area on the surface of the pine tree and a turpentine volume in the turpentine existing area; The turpentine estimation model and the turpentine volume in the turpentine existing area are used to select at least one nematode affected area from the turpentine existing area; The at least one pest situation area where the pine wood nematode exists in the feeding area is identified according to the distribution of the nematode affected area; When the number of the pest situation areas is 0, no pre-warning is given, otherwise, the pest situation areas in the pine tree are pre-warned.

Images