A peanut pest monitoring and management method and planting management system

Abstract

The present invention discloses a peanut pest monitoring and management method and planting management system, relating to the field of agricultural management technology. The present invention comprises evenly dividing a monitoring and management range into multiple monitoring grids, selecting peanut plants as sample plants within each monitoring grid; identifying the pest type of the sample plants at each stage of peanut growth to obtain the probability that each sample plant belongs to each pest type; analyzing the distribution area of ​​each pest type within the monitoring and management range based on the location of the sample plant in the monitoring grid and the probability that each sample plant belongs to each pest type; obtaining the pest type of each peanut plant based on the distribution area of ​​each pest type within the monitoring and management range and the location of each peanut plant; determining whether the pest type of the peanut plant is pest-free; if so, not performing any treatment; if not, performing field pest control according to the corresponding pest disposal plan. The present invention improves the accuracy and efficiency of pest management for planting.

Description

Technical Field

[0001] The invention belongs to the technical field of agricultural management, and in particular relates to a peanut pest monitoring and management method and a planting management system. Background Art

[0002] Peanuts are extremely vulnerable to a variety of pests during their growth process, including underground pests such as white grubs and cutworms, as well as aboveground pests such as aphids and spider mites. These pests not only directly damage the roots, stems, leaves, and fruit of peanuts, leading to reduced or even complete crop failure, but also spread diseases, further exacerbating losses.

[0003] Traditional peanut pest monitoring mainly relies on manual field surveys, which have problems such as low efficiency, poor timeliness, and low accuracy, and it is difficult to meet the needs of modern precision agriculture. Summary of the Invention

[0004] The purpose of the present invention is to provide a peanut pest monitoring and management method and a planting management system to improve the accuracy and efficiency of pest planting management.

[0005] To solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention provides a peanut pest monitoring and management method, comprising:

[0007] The monitoring management area is evenly divided into multiple monitoring grids, and a peanut plant is selected as a sample plant in each monitoring grid;

[0008] Obtain the pest type identification results of sample plants at each stage of peanut growth and obtain the probability that each sample plant belongs to each pest type;

[0009] The distribution area of ​​each pest type within the monitoring management scope is obtained based on the location of each sample plant in the monitoring grid and the probability that each sample plant belongs to each pest type;

[0010] The pest type of each peanut plant is obtained based on the distribution area of ​​each pest type and the location of each peanut plant within the monitoring and management scope;

[0011] Determine whether the pest type of peanut plants is pest-free;

[0012] If so, no processing is performed;

[0013] If not, carry out field pest control according to the corresponding pest management plan.

[0014] The present invention also discloses a peanut pest monitoring and planting management system, comprising:

[0015] An identification and entry unit is used for allowing each staff member to enter the identified pest type of each sample plant after identifying the pest type of several parts of the sample plant;

[0016] A monitoring and management unit is used to evenly divide the monitoring and management range into a plurality of monitoring grids, and select a peanut plant in each monitoring grid as a sample plant;

[0017] Obtain the pest type identification results of sample plants at each stage of peanut growth and obtain the probability that each sample plant belongs to each pest type;

[0018] The distribution area of ​​each pest type within the monitoring management scope is obtained based on the location of each sample plant in the monitoring grid and the probability that each sample plant belongs to each pest type;

[0019] The pest type of each peanut plant is obtained based on the distribution area of ​​each pest type and the location of each peanut plant within the monitoring and management scope;

[0020] A control management unit is used to determine whether the pest type of the peanut plant is pest-free;

[0021] If so, no processing is performed;

[0022] If not, carry out field pest control according to the corresponding pest management plan.

[0023] The present invention obtains the accurate distribution area of ​​each pest type within the monitoring and management range by analyzing and comparing the preliminary pest identification status of a small number of sample plants by the monitoring and management unit, thereby realizing the automatic and rapid detection and identification of all peanut plant pest types within the monitoring and management range, and fully improving the efficiency of pest planting management accuracy.

[0024] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 This is a schematic diagram of the functional units and information flow of a peanut pest monitoring and planting management system according to one embodiment of the present invention;

[0027] Figure 2A schematic diagram of a process flow of the monitoring management unit and the governance management unit according to an embodiment of the present invention;

[0028] Figure 3 This is a schematic diagram of the process flow of step S2 in one embodiment of the present invention;

[0029] Figure 4 This is a schematic diagram of the process flow of step S3 in one embodiment of the present invention;

[0030] Figure 5 This is a schematic diagram of the process flow of step S31 in one embodiment of the present invention;

[0031] Figure 6 This is a schematic diagram of the process flow of step S32 in one embodiment of the present invention;

[0032] Figure 7 This is a schematic diagram of the process flow of step S4 in one embodiment of the present invention;

[0033] Figure 8 This is a schematic diagram of the process flow of step S42 in one embodiment of the present invention;

[0034] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0035] 1-Identification and entry unit, 2-Monitoring and management unit, 3-Governance and management unit. DETAILED DESCRIPTION

[0036] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0037] It should be noted that the terms "first," "second," and the like in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are merely examples of devices and methods consistent with certain aspects of the present application as detailed in the appended claims.

[0038] See also Figures 1 to 2As shown, the present invention provides a peanut pest monitoring and planting management system, which is divided into functional units including an identification and entry unit 1, a monitoring and management unit 2, and a governance and management unit 3. The identification and entry unit 1 in this system can be a hardware entity or an application installed on a mobile phone or computer. It is used in step S01 for a staff member to enter the identified pest type of each sample plant. Alternatively, the staff member can manually identify the pest type of the sample plant and then enter the pest type.

[0039] The parts of peanut plants include stems, leaves and flowers. The stems, leaves and flowers of peanut plants are visible parts, and there are also buried parts such as roots, fruits, and seeds. In the process of identifying pests and diseases of peanut plants, the results of identifying different parts of different plants may be different, and the results of visual identification or identification by different staff members using means such as intelligent comparison and genetic testing may also be different. The identification and entry unit 1 can execute step S01 to obtain the pest type of the sample plant by the staff. The insect types include no pests, peanut aphids, leaf rollers, tea yellow mites, beet armyworms, field spider mites, white grubs, leafworms, cotton bollworms, thrips and / or peanut spider scales.

[0040] The techniques used by pest identification workers are becoming increasingly diverse, combining traditional methods with modern technology. Visual inspection can be based on empirical observation of pest morphology (color, size, antennae, etc.) and symptoms of damage (holes, excrement, leaf defects, etc.), sometimes combined with microscopic identification. Alternatively, digital intelligent identification technology can be used, such as taking photos of pests and automatically comparing them to databases using AI algorithms, such as those used by iNaturalist, Plantix, and Google Lens. Other methods include the "Shi Nong" app and Tencent AI Lab's pest and disease identification tools. Molecular biology techniques can also be used to extract pest DNA fragments (such as the COI gene) and compare them to databases for precise species identification.

[0041] Visual inspections and digital intelligent identification typically consider the damage patterns caused by different pests to peanut plants. For example, when peanuts are infested by peanut aphids, the leaves turn yellow and curl due to the aphids' sap-sucking, which also affects stem growth and flower development. When peanuts are infested by leaf rollers, the larvae spin silk to wrap the leaves into buds and feed on the mesophyll, causing them to shrink or fall. In severe cases, the upper leaves turn white, affecting flower and pod formation. When peanuts are infested by tea yellow mites, adult and larvae concentrate on young shoots, tender leaves, flowers, and young fruits to feed on sap, causing thickening, stiffening, and curling of leaf margins. In severe cases, this can affect flowering and fruiting. When peanuts are infested by beet armyworms, the larvae feed on the leaves, causing notches or holes. In severe cases, the entire leaf is consumed, leaving only the petiole and veins, reducing yield. When peanuts are infested by spider mites, the main damage is to the quality and yield of peanuts, and the sap is sucked, causing poor plant growth. When peanuts are infested by white grubs, the larvae damage the fruit needles and young fruits, causing the plants to be stunted, the leaves to turn yellow, and in severe cases, the plants to die. When peanuts are infested by the armyworm, the larvae feed on the leaves, causing notches or holes in the leaves, and in severe cases, eating up all the leaves. When peanuts are infested by cotton bollworms, the larvae damage the leaves and pods of peanuts, affecting yield. When peanuts are infested by thrips, they suck the sap from the young leaves and inflorescences of peanuts, causing leaf deformation and poor growth. When peanuts are infested by the new peanut spider scale, the larvae drill into the root bark to suck the sap, causing the plants to be stunted, the leaves to turn yellow, and in severe cases, the plants to die.

[0042] Please continue reading Figures 1 to 3 As shown, since the results of pest identification on different parts of the peanut plant by different staff members may be different during the operation of the identification and entry unit 1, after executing step S21 to obtain the pest type of the sample plant obtained by each staff member through pest identification on each part of the sample plant, it is also necessary to execute step S22 to statistically obtain the proportion of the sample plant belonging to each pest type as the probability that the sample plant belongs to each pest type.

[0043] Please continue reading Figures 4 to 5As shown, after obtaining the preliminary results of pest identification of the identification entry unit 1, since the pest identification results of a single peanut plant, that is, the preliminary identification results, may not be completely accurate, it is necessary to combine the position of each sample plant and the identification results for auxiliary judgment. Therefore, step S3 can be executed next to obtain the distribution area of ​​each pest type within the monitoring and management scope based on the position of the monitoring grid where each sample plant is located and the probability analysis of each sample plant belonging to each pest type. Specifically, step S31 can be executed first to select a reference sample plant from all the sample plants. Specifically, step S311 can be executed first to take the pest type to which all the sample plants belong whose probability is not 0 as a suspected pest type. Next, step S312 can be executed for each suspected pest type, and the sample plant corresponding to the maximum probability value of the pest type among all the sample plants can be used as a reference sample plant.

[0044] See also Figures 2 to 6 As shown, since pests are contagious, pests on peanut plants usually occur in clusters. Although the pest identification results of a single peanut plant may not be accurate, the summary analysis of the identification results of many peanut plants is reliable. Therefore, the distribution area of ​​each pest type can be assisted by the location distribution of the plants. Therefore, step S32 can be executed next to obtain the distribution area of ​​each pest type within the detection and management range based on the difference in the probability of each pest type between the reference sample plant and other sample plants. Specifically, step S321 can be executed first to take the cumulative value of the absolute value of the probability difference on each pest type between the sample plants as the pest identification difference between the sample plants, and calculate the pest identification difference between each reference sample plant and each other sample plant. Next, step S322 can be executed to divide the other sample plants other than each reference sample plant into the same distribution area with the reference sample plant with the smallest pest identification difference. Next, step S323 can be executed to calculate the average probability of each pest type for all sample plants in each distribution area. Next, step S324 may be executed to take the several pest types with average probabilities greater than the set warning value as the pest types of the distribution area, and obtain the distribution areas of the various pest types within the detection and management range.

[0045] Please continue reading Figure 5 and 6As shown, the pest types of the plants in the distribution areas divided in the above steps may not be consistent, so it is necessary to perform an inspection and judgment. During the inspection process, step S325 can be first executed to determine whether the sample plant with the smallest difference in pest identification between each distribution area and the average probability of each pest type of all sample plants in the distribution area is the reference sample plant in the distribution area. If so, it means that the pest types of the peanut plants in each distribution area are consistent, so step S326 can be executed next to obtain the distribution areas of each pest type within the detection and management scope. If not, it means that the pest types of the peanut plants in the distribution area are inconsistent, so step S327 can be executed next to reselect the reference sample plant; calculate and obtain the sample plant with the smallest difference in pest identification between each distribution area and the average probability of each pest type of all sample plants in the distribution area as the reselected reference sample plant. In order to ensure that the pest types of peanut plants in each distribution area are consistent, the distribution area division method can be continuously optimized through iterative calculation and judgment. That is, returning to step S322 to step S325, the updated distribution area is recalculated and divided based on the newly selected reference sample plants, and re-determining whether the sample plant with the smallest pest identification difference between each distribution area and the average probability of each pest type for all sample plants in the distribution area is the reference sample plant in the distribution area. The above process is continuously iteratively optimized to obtain distribution areas containing peanut plants with consistent pest types.

[0046] To supplement the implementation of steps S321 to S327, we provide the source code for some functional modules, with cross-references and explanations provided in the comments. To prevent the leakage of data involving commercial secrets, data that does not affect the implementation of the solution is desensitized. The same applies below.

[0047] #include <iostream>

[0048] #include <vector>

[0049] #include <map>

[0050] #include <cmath>

[0051] #include <algorithm>

[0052] #include <limits>

[0053] using namespace std;

[0054] / / Define pest type enumeration

[0055] enum PestType {

[0056] NO_PEST, / / No pests

[0057] APHID, / / Peanut Aphid

[0058] LEAF_ROLLER, / / Leaf Roller

[0059] TEA_MITE, / / Tea yellow mite

[0060] BEET_ARMYWORM, / / Beet Armyworm

[0061] FIELD_MITE, / / Field spider mite

[0062] GRUB, / / Grub

[0063] SPODOPTERA, / / Spodoptera litura

[0064] BOLLWORM, / / bollworm

[0065] THRIPS, / / thrips

[0066] NEW_SPIDER_MITE / / New Spider Mite

[0067] };

[0068] / / Define the sample plant structure

[0069] struct TargetPlant {

[0070] int id; / / Plant ID

[0071] map<PestType, double> pestProbs; / / probability of each pest type

[0072] };

[0073] / / Calculate the difference in pest identification

[0074] double calculatePestDifference(const map<PestType, double>& probs1,const map<PestType, double>& probs2) {

[0075] double difference = 0.0;

[0076] for (const auto& prob : probs1) {

[0077] difference += abs(prob.second - probs2.at(prob.first));

[0078] }

[0079] return difference;

[0080] }

[0081] / / Divide the distribution area

[0082] map<int, vector <targetplant>> divideDistributionRegions(const vector <targetplant>& plants, const vector <targetplant>& landmarks) {

[0083] map<int, vector <targetplant>> regions;

[0084] / / Traverse all sample plants

[0085] for (const auto& plant : plants) {

[0086] double minDifference = numeric_limits <double>::max();

[0087] int closestLandmarkId = -1;

[0088] / / Find the reference sample plant with the smallest difference in pest identification from the current plant

[0089] for (const auto& landmark : landmarks) {

[0090] double difference = calculatePestDifference(plant.pestProbs, landmark.pestProbs);

[0091] if (difference < minDifference) {

[0092] minDifference = difference;

[0093] closestLandmarkId = landmark.id;

[0094] }

[0095] }

[0096] / / Divide the current plant into the corresponding distribution area

[0097] regions[closestLandmarkId].push_back(plant);

[0098] }

[0099] return regions;

[0100] }

[0101] / / Calculate the average probability of the distribution area

[0102] map<PestType, double> calculateAverageProbabilities(const vector <targetplant>& plants) {

[0103] map<PestType, double> avgProbs;

[0104] map<PestType, int> count;

[0105] / / Initialize average probability and count

[0106] for (const auto& plant : plants) {

[0107] for (const auto& prob : plant.pestProbs) {

[0108] avgProbs[prob.first] += prob.second;

[0109] count[prob.first]++;

[0110] }

[0111] }

[0112] / / Calculate the average probability

[0113] for (auto& prob : avgProbs) {

[0114] prob.second / = count[prob.first];

[0115] }

[0116] return avgProbs;

[0117] }

[0118] / / Determine whether the reference sample plant needs to be updated

[0119] bool needUpdateLandmarks(const map <int, vector <targetplant>>®ions, const vector <targetplant>& landmarks) {

[0120] for (const auto& region : regions) {

[0121] int landmarkId = region.first;

[0122] const auto& plants = region.second;

[0123] / / Calculate the average probability of the current distribution area

[0124] auto avgProbs = calculateAverageProbabilities(plants);

[0125] / / Find the sample plant with the smallest difference from the average probability

[0126] double minDifference = numeric_limits <double>::max();

[0127] int closestPlantId = -1;

[0128] for (const auto& plant : plants) {

[0129] double difference = calculatePestDifference(plant.pestProbs, avgProbs);

[0130] if (difference < minDifference) {

[0131] minDifference = difference;

[0132] closestPlantId = plant.id;

[0133] }

[0134] }

[0135] / / If the sample plant with the smallest difference is not the current reference sample plant, it needs to be updated

[0136] if (closestPlantId != landmarkId) {

[0137] return true;

[0138] }

[0139] }

[0140] return false;

[0141] }

[0142] / / Update reference sample plants

[0143] vector <targetplant>updateLandmarks(const map<int, vector <targetplant>>& regions) {

[0144] vector <targetplant>newLandmarks;

[0145] for (const auto& region : regions) {

[0146] const auto& plants = region.second;

[0147] / / Calculate the average probability of the current distribution area

[0148] auto avgProbs = calculateAverageProbabilities(plants);

[0149] / / Find the sample plant with the smallest difference from the average probability

[0150] double minDifference = numeric_limits <double>::max();

[0151] TargetPlant closestPlant;

[0152] for (const auto& plant : plants) {

[0153] double difference = calculatePestDifference(plant.pestProbs, avgProbs);

[0154] if (difference < minDifference) {

[0155] minDifference = difference;

[0156] closestPlant = plant;

[0157] }

[0158] }

[0159] / / Use the sample plant with the smallest difference as the new reference sample plant

[0160] newLandmarks.push_back(closestPlant);

[0161] }

[0162] return newLandmarks;

[0163] }

[0164] int main() {

[0165] / / 1. The example has obtained sample plant data and reference sample plants

[0166] vector <targetplant>targetPlants;

[0167] for (int i = 0; i < 10; ++i) {

[0168] TargetPlant plant;

[0169] plant.id = i;

[0170] plant.pestProbs[APHID] = static_cast <double>(rand()) / RAND_MAX;

[0171] plant.pestProbs[LEAF_ROLLER] = static_cast <double>(rand()) / RAND_MAX;

[0172] plant.pestProbs[TEA_MITE] = static_cast <double>(rand()) / RAND_MAX;

[0173] targetPlants.push_back(plant);

[0174] }

[0175] vector <targetplant>landmarks = {targetPlants[0], targetPlants[5]}; / / Example reference sample plants

[0176] / / 2. Divide the initial distribution area

[0177] auto regions = divideDistributionRegions(targetPlants,landmarks);

[0178] / / 3. Iteratively optimize the distribution area

[0179] while (needUpdateLandmarks(regions, landmarks)) {

[0180] landmarks = updateLandmarks(regions);

[0181] regions = divideDistributionRegions(targetPlants, landmarks);

[0182] }

[0183] / / 4. Output the final distribution area

[0184] cout << "Final distribution area division result:" << endl;

[0185] for (const auto& region : regions) {

[0186] cout << "Reference sample plant ID: " << region.first << ", including plant ID:";

[0187] for (const auto& plant : region.second) {

[0188] cout << plant.id << " ";

[0189] }

[0190] cout << endl;

[0191] }

[0192] return 0;

[0193] }

[0194] Function summary:

[0195] This code implements the function of dividing the pest distribution area according to the difference in pest identification between the reference sample plant and other sample plants, and optimizing the distribution area through iteration. First, the pest identification difference is calculated, and the cumulative value of the absolute value of the difference in the probability of pest types between the sample plants is calculated as the pest identification difference. Then, the distribution area is divided, and each sample plant is divided into the distribution area where the reference sample plant with the smallest pest identification difference is located. Next, the average probability is calculated, and the average probability of the sample plants in each distribution area for each pest type is calculated. Then, it is determined whether the reference sample plants need to be updated. If the sample plant with the smallest difference from the average probability in a certain distribution area is not the current reference sample plant, the reference sample plant needs to be updated. Finally, the distribution area is optimized iteratively by updating the reference sample plants and re-dividing the distribution area until the distribution area is stable.

[0196] The above code is developed through a detailed algorithm to achieve dynamic division and optimization of pest distribution areas, providing a scientific basis for pest monitoring and management.

[0197] See also Figure 2 、 7 As shown in FIG8 , since the accuracy of the output result of the identification and entry unit 1 may not meet the requirements, an auxiliary determination can be made based on the pest type in the distribution area of ​​the peanut plant. Specifically, step S4 is executed to determine the pest type of each peanut plant based on the distribution area of ​​each pest type within the monitoring and management area and the location of each peanut plant. Specifically, step S41 can be first executed to obtain the set pest identification accuracy. Next, step S42 can be executed to determine whether the probability of each sample plant for each pest type meets the set pest identification accuracy based on the probability of each pest type. In step S42, when determining whether the identification and entry unit 1 is sufficiently accurate, step S421 can first be executed to calculate the difference between the maximum probability and the second maximum probability for each pest type of the sample plant as the pest type probability gradient value of the sample plant. Next, step S422 can be executed to calculate the ratio of the pest type probability gradient value to the maximum probability of the sample plant as the pest identification accuracy of the sample plant. Then, step S423 can be executed to determine whether the pest identification accuracy of the sample plant is greater than or equal to the pest identification accuracy. If yes, step S424 can be executed to determine whether the probability of the sample plant on each pest type meets the set accuracy of pest identification. If no, step S425 can be executed to determine whether the probability of the sample plant on each pest type meets the set accuracy of pest identification.

[0198] If the probability of the sample plant for each pest type is determined to meet the set accuracy for pest identification, then the pest identification result is directly usable, and therefore step S43 can be executed to output the pest type that meets the set accuracy for pest identification. If the probability of the sample plant for each pest type is determined to be inconsistent with the set accuracy for pest identification, then the pest identification result is unreliable, and therefore the pest status of the distribution area in which the sample plant is located can be used as the pest status of the peanut plant. That is, step S44 is executed to use the pest type corresponding to the distribution area in which the sample plant's location falls as the pest type of the sample plant.

[0199] Please continue reading Figures 1 to 3 As shown, the control management unit 3 in this system can be an intelligent terminal guided by the monitoring management unit 2, which is received by the field management personnel and then operated. During operation, step S5 can be first executed to determine whether the pest type of the peanut plant is pest-free. If so, step S5 can be executed without processing. Otherwise, step S7 can be executed next to carry out field pest control according to the corresponding pest disposal plan. The specific means can be to use lights, color plates, etc. to lure and kill pests, or to select high-efficiency and low-toxic pesticides, such as imidacloprid and cypermethrin, to kill pests by spraying, root irrigation, etc.

[0200] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, systems, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a part for a module, program segment or instruction, and the part for the module, program segment or instruction comprises one or more executable instructions for realizing the logical function of the specification. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous boxes can actually be performed substantially in parallel, and they can sometimes also be performed in the opposite order, depending on the function involved.

[0201] It should also be noted that each box in the block diagram and / or flowchart, and combinations of boxes in the block diagram and / or flowchart, can be implemented by hardware that performs the corresponding function or action, such as a circuit or ASIC (Application Specific Integrated Circuit), or can be implemented by a combination of hardware and software, such as firmware.

[0202] Although the present invention has been described herein in conjunction with various embodiments, in the process of implementing the claimed invention, those skilled in the art may understand and implement other variations of the disclosed embodiments by examining the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple situations. A single processor or other unit may implement several functions listed in the claims. The fact that certain measures are recorded in different dependent claims does not mean that these measures cannot be combined to produce good results.

[0203] The embodiments of the present application have been described above. The above description is illustrative and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein.< / targetplant> < / double> < / double> < / double> < / targetplant> < / double> < / targetplant> < / targetplant> < / targetplant> < / double> < / targetplant> < / targetplant> < / targetplant> < / double> < / targetplant> < / targetplant> < / targetplant> < / targetplant> < / limits> < / algorithm> < / cmath> < / map> < / vector> < / iostream>

Claims

1. A peanut pest monitoring and management method, characterized in that: include, The monitoring management area is evenly divided into multiple monitoring grids, and a peanut plant is selected as a sample plant in each monitoring grid; Obtain the pest type identification results of sample plants at each stage of peanut growth and obtain the probability that each sample plant belongs to each pest type; Among the pest types to which all sample plants belong, the pest types with a probability not equal to 0 are regarded as suspected pest types; For each suspected pest type, the sample plant with the highest probability value of that pest type among all sample plants is used as the reference sample plant; The distribution areas of each pest type within the detection and management scope were obtained based on the difference in the probability of each pest type between the reference sample plants and other sample plants; Get the accuracy of pest identification settings; Determine whether the accuracy of pest identification is met based on the probability of each sample plant for each pest type; If so, the pest type corresponding to the maximum probability of the sample plant in each pest type is used as the pest type of all peanut plants in the monitoring grid where the sample plant is located; If not, the pest type of the distribution area where the sample plant is located is used as the pest type of all peanut plants in the monitoring grid where the sample plant is located; Determine whether the pest type of peanut plants is pest-free; If so, no processing is performed; If not, carry out field pest control according to the corresponding pest management plan; The step of obtaining the distribution area of ​​each pest type within the detection management range based on the difference in the probability of each pest type between the reference sample plant and other sample plants includes: The cumulative value of the absolute value of the probability difference of each pest type between the sample plants is used as the pest identification difference between the sample plants, and the pest identification difference between each reference sample plant and each other sample plant is calculated; The sample plants other than each reference sample plant and the reference sample plant with the smallest pest identification difference are divided into the same distribution area, wherein the reference sample plant with the smallest pest identification difference is the reference sample plant with the smallest pest identification difference between the other sample plants and each reference sample plant; Calculate the average probability of each pest type for all sample plants in each distribution area; Several pest types with an average probability greater than a set warning value are regarded as the pest types in the distribution area to obtain the distribution areas of each pest type within the detection and management scope.

2. The method according to claim 1, characterized in that The step of obtaining the pest type identification results of the sample plants at each stage of peanut growth to obtain the probability that each sample plant belongs to each pest type, include, Obtaining the pest type of the sample plant obtained by performing pest identification on each part of the sample plant each time, wherein the parts of the peanut plant include stems, leaves, and / or flowers; The proportion of sample plants belonging to each pest type is obtained as the probability that the sample plants belong to each pest type, wherein the pest types include no pests, peanut aphids, leaf rollers, tea yellow mites, beet armyworms, field spider mites, white grubs, Spodoptera litura, cotton bollworms, thrips and / or peanut spider scales.

3. The method according to claim 1, characterized in that The step of obtaining the distribution area of ​​each pest type within the detection management range based on the difference in the probability of each pest type between the reference sample plant and other sample plants also includes: Determine whether the sample plant with the smallest pest identification difference between the average probability of each pest type for all sample plants in the distribution area in each distribution area is the reference sample plant in the distribution area; If so, the distribution areas of each pest type within the detection and management scope are obtained; If not, reselect reference sample plants; Recalculate and divide the updated distribution area based on the reselected reference sample plants; Re-judge whether the sample plant with the smallest pest identification difference between the average probability of each pest type of all sample plants in the distribution area is the reference sample plant in the distribution area.

4. The method according to claim 3, characterized in that The step of reselecting reference sample plants includes: The sample plant with the smallest difference in pest identification between the average probability of each pest type for all sample plants in each distribution area is calculated and obtained as the reference sample plant for reselection.

5. The method according to claim 1, wherein The step of judging whether the probability of each sample plant for each pest type meets the set accuracy of the pest identification includes: Calculate the difference between the maximum probability and the second maximum probability of each pest type of the sample plant as the probability gradient value of the pest type of the sample plant; The ratio of the probability gradient value of the pest type of the sample plant to the maximum probability is used as the pest identification accuracy of the sample plant; Determining whether the pest identification accuracy of the sample plant is greater than or equal to the pest identification set accuracy; If so, the probability of the sample plant being on each pest type is determined to meet the set accuracy of the pest identification; If not, then vice versa.

6. A peanut pest monitoring and planting management system, characterized in that: include, An identification and entry unit is used to enter the identified pest type of each sample plant after identifying the pest type of several parts of the sample plant; A monitoring and management unit is used to evenly divide the monitoring and management range into a plurality of monitoring grids, and select a peanut plant in each monitoring grid as a sample plant; Obtain pest type identification results of sample plants at each stage of peanut growth and obtain the probability that each sample plant belongs to each pest type; Among the pest types to which all sample plants belong, the pest types with a probability not equal to 0 are regarded as suspected pest types; For each suspected pest type, the sample plant with the highest probability value of that pest type among all sample plants is used as the reference sample plant; The distribution areas of each pest type within the detection and management scope were obtained based on the difference in the probability of each pest type between the reference sample plants and other sample plants; Get the accuracy of pest identification settings; Determine whether the accuracy of pest identification is met based on the probability of each sample plant for each pest type; If so, the pest type corresponding to the maximum probability of the sample plant in each pest type is used as the pest type of all peanut plants in the monitoring grid where the sample plant is located; If not, the pest type of the distribution area where the sample plant is located is used as the pest type of all peanut plants in the monitoring grid where the sample plant is located; The step of obtaining the distribution area of ​​each pest type within the detection management range based on the difference in the probability of each pest type between the reference sample plant and other sample plants includes: The cumulative value of the absolute value of the probability difference of each pest type between the sample plants is used as the pest identification difference between the sample plants, and the pest identification difference between each reference sample plant and each other sample plant is calculated; The sample plants other than each reference sample plant and the reference sample plant with the smallest pest identification difference are divided into the same distribution area, wherein the reference sample plant with the smallest pest identification difference is the reference sample plant with the smallest pest identification difference between the other sample plants and each reference sample plant; Calculate the average probability of each pest type for all sample plants in each distribution area; The several pest types with average probabilities greater than the set warning value are all regarded as the pest types in the distribution area to obtain the distribution areas of each pest type within the detection and management scope; A control management unit is used to determine whether the pest type of the peanut plant is pest-free; If so, no processing is performed; If not, carry out field pest control according to the corresponding pest management plan.

Images