Method for detecting rice blast of rice leaves based on monitoring image

Abstract

The invention discloses a rice leaf rice blast detection method based on a monitoring image, relates to the field of agricultural planting, solves the problem of poor detection effect of the existing rice leaf rice blast detection method, and comprises the following steps: S1, performing periodic image acquisition on a target to-be-detected farmland to obtain a plurality of rice leaf image sets, the method comprises the following steps: S1, respectively carrying out leaf image rice blast analysis on rice plants in each rice leaf image set, and obtaining leaf image preliminary analysis data according to an analysis result, S2, carrying out batch disease blast plant marking in a farmland region coordinate system according to the leaf image preliminary analysis data to obtain disease blast rice image marking data, and S3, performing disease blast plant stage evolution analysis on the target to-be-detected farmland according to the disease blast rice image marking data, and issuing farmland disease blast early warning according to an analysis result. The detection method for the rice blast of the rice leaves is comprehensive and accurate.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of agricultural planting, and relates to image monitoring technology, and particularly relates to a rice leaf rice blast detection method based on monitoring images. BACKGROUND

[0002] The existing rice leaf blast detection method has the following defects when detecting the blast in farmland:

[0003] 1. The existing rice leaf blast detection method can only analyze the blast images of single rice leaves to give a blast warning for the single rice, and cannot comprehensively analyze the rice blast images in the farmland, so that the blast detection result lacks comprehensiveness.

[0004] 2. The existing rice leaf blast detection method cannot create a plane coordinate system for the target farmland to be detected, and cannot collect rice images at different stages to mark the blast plants in the farmland area coordinate system in batches, so that the blast plant stage evolution analysis cannot be carried out on the target farmland to be detected according to the marking result, and the blast warning cannot be given according to the analysis result, and the detection method is single and has poor application promotion.

[0005] Therefore, the application provides a rice leaf blast detection method based on monitoring images. SUMMARY

[0006] In view of the defects of the prior art, the application aims to provide a rice leaf blast detection method based on monitoring images, and aims to improve the comprehensiveness and application promotion of the rice leaf blast detection method.

[0007] In order to achieve the above-mentioned purpose, the application adopts the following technical scheme:

[0008] Step S1: obtaining the target farmland to be detected, periodically acquiring images of the target farmland to be detected, obtaining a plurality of rice leaf image sets, and analyzing the rice plants in each rice leaf image set to obtain preliminary analysis data of the leaf images according to the analysis result;

[0009] Step S2: creating a plane coordinate system for the target farmland to be detected, obtaining a farmland area coordinate system, and marking the blast plants in the farmland area coordinate system in batches according to the preliminary analysis data of the leaf images to obtain blast rice image marking data;

[0010] Step S3: performing blast plant stage evolution analysis on the target farmland to be detected according to the blast rice image marking data, and giving a blast warning for the farmland according to the analysis result.

[0011] Further, the step S1 further includes the following steps.

[0012] Step S11: Obtain the rice fields in need of rice blast detection, and randomly select one target field to be detected from the obtained multiple rice fields;

[0013] Step S12: In the process of detecting the target field to be detected, mark a rice disease and pest detection cycle with the time point corresponding to the current time as the cycle end time point;

[0014] Step S13: Set a plurality of image acquisition batches in the rice disease and pest detection cycle, and name the set image acquisition batches according to the order of acquisition time as P1 image acquisition batch to Pa image acquisition batch;

[0015] Step S14: Obtain the rice leaf images from the P1 image acquisition batch of the target field to be detected to form a P1 rice leaf image set, obtain the rice leaf images from the P2 image acquisition batch of the target field to be detected to form a P2 rice leaf image set, and so on, and obtain the rice leaf images from the Pa image acquisition batch of the target field to be detected to form a Pa rice leaf image set;

[0016] Step S15: Perform rice leaf blast analysis on the P1 rice leaf image set, obtain the plant blast index corresponding to each rice plant according to the analysis result, and obtain P1 leaf blast analysis data;

[0017] Step S16: Obtain the leaf blast analysis data corresponding to the P2 rice leaf image set to the Pa rice leaf image set, and obtain P2 leaf blast analysis data to Pa leaf blast analysis data;

[0018] Step S17: Define the P1 leaf blast analysis data to the Pa leaf blast analysis data as leaf image preliminary analysis data.

[0019] Further, the step S15 further includes the following steps.

[0020] Step S151: Obtain the rice plants planted in the target field to be detected, and randomly select one as a sample rice plant from the obtained multiple rice plants;

[0021] Step S152: Obtain the rice leaf images corresponding to the sample rice plant, obtain a plurality of rice leaf images, and select one sample rice leaf image from the obtained multiple rice leaf images;

[0022] Step S153: performing leaf blight analysis on the sample rice leaf image, and obtaining an image blight index corresponding to the sample rice leaf image according to an analysis result;

[0023] Step S154: obtaining an image blight index corresponding to each rice leaf image respectively, and performing mean calculation on the obtained multiple image blight indexes to obtain a plant blight index corresponding to the sample rice plant;

[0024] Step S155: obtaining the plant blight index corresponding to each rice plant respectively to obtain P1 leaf blight analysis data.

[0025] Further, the step S153 further includes the following steps:

[0026] marking a rice leaf coverage area in the sample rice leaf image using an image recognition algorithm to obtain an image leaf area;

[0027] segmenting the image leaf area into a plurality of leaf pixel points, and using an RGB color model to obtain a color R value, a color G value and a color B value corresponding to each leaf pixel point respectively;

[0028] obtaining a plurality of rice leaf historical images, and the leaf blight pixel points and the healthy leaf pixel points in the rice leaf historical images have been marked;

[0029] obtaining a color R value interval, a color G value interval and a color B value interval corresponding to the leaf blight pixel points to obtain a first R value preset interval, a first G value preset interval and a first G value preset interval;

[0030] obtaining a color R value interval, a color G value interval and a color B value interval corresponding to the healthy leaf pixel points to obtain a second R value preset interval, a second G value preset interval and a second G value preset interval;

[0031] obtaining leaf pixel points with the color R value, the color G value and the color B value in the first R value preset interval, the first G value preset interval and the first B value preset interval respectively to obtain a plurality of blight area pixel points;

[0032] obtaining a deviation of the color R value, the color G value and the color B value corresponding to each blight area pixel point from the second R value preset interval, the second G value preset interval and the second G value preset interval respectively to obtain a color R value deviation, a color G value deviation and a color B value deviation corresponding to each blight area pixel point;

[0033] performing mean calculation on the color R value deviation, the color G value deviation and the color B value deviation corresponding to the same blight area pixel point to obtain a plurality of color RGB deviations;

[0034] counting the number of pixels in the image leaf area to obtain a first pixel number value, and counting the number of pixels in the image leaf area to obtain a second pixel number value;

[0035] calculating the image rice leaf blight index corresponding to the sample rice leaf image by using the color RGB deviation, the first pixel number value and the second pixel number value;

[0036] calculating the image rice leaf blight index corresponding to the sample rice leaf image, and the specific formula is as follows:

[0037]

[0038] Wherein, Wbz is the image rice leaf blight index corresponding to the sample rice leaf image, Xsz1 is the first pixel number value, Xsz2 is the second pixel number value, and Yxs is the color RGB deviation.

[0039] Further, the step S2 further includes the following specific steps:

[0040] Step S21: creating a coordinate system for the target farmland to be detected to obtain a target farmland plane coordinate system;

[0041] Step S22: obtaining leaf image preliminary analysis data, and obtaining P1 leaf blight analysis data to Pa leaf blight analysis data according to the leaf image preliminary analysis data;

[0042] Step S23: judging the growth state type of each rice plant according to the P1 leaf blight analysis data, and marking the judgment result in the target farmland plane coordinate system to obtain a P1 stage farmland marking image;

[0043] Step S24: respectively classifying and marking the coordinates of the rice plants in the P2 leaf blight analysis data to the Pa leaf blight analysis data to obtain a P2 stage farmland marking image to a Pa stage farmland marking image, and obtaining the leaf blight image marking data;

[0044] The step S23 further includes the following specific steps:

[0045] According to the P1 leaf blight analysis data, obtaining the plant blight index corresponding to each rice plant in the P1 image collection batch according to the P1 leaf blight analysis data;

[0046] If the plant blight index is 0, the corresponding rice plant is marked as a first type of rice plant, and if the plant blight index is greater than 0, the corresponding rice plant is marked as a second type of rice plant;

[0047] The first type of rice plant and the second type of rice plant are marked in the target farmland plane coordinate system respectively in the target farmland plane image, to obtain a P1 stage farmland marking image.

[0048] Further, the step S3 further includes the following steps:

[0049] Step S31: Obtain the rice blast image marking data, and obtain P1 stage farmland marking image to Pa stage farmland marking image according to the rice blast image marking data respectively;

[0050] Step S32: Perform rice blast area analysis on the P1 stage farmland marking image, and obtain a P1 image rice blast area ratio according to the analysis result;

[0051] Step S33: Obtain the image rice blast area ratio corresponding to the P2 stage farmland marking image to the Pa stage farmland marking image respectively, to obtain a P2 image rice blast area ratio to a Pa image rice blast area ratio;

[0052] Step S34: Calculate the difference between the P2 image rice blast area ratio and the P1 image rice blast area ratio, and calculate the ratio of the obtained difference to the P1 image rice blast area ratio, to obtain a K1 rice blast area change rate; calculate the difference between the P3 image rice blast area ratio and the P2 image rice blast area ratio, and calculate the ratio of the obtained difference to the P2 image rice blast area ratio, to obtain a K2 rice blast area change rate; and so on, to calculate the difference between the Pa image rice blast area ratio and the Pa-1 image rice blast area ratio, and calculate the ratio of the obtained difference to the Pa-1 image rice blast area ratio, to obtain a Kd rice blast area change rate;

[0053] Step S35: Calculate the average of the K1 rice blast area change rate and the Kd rice blast area change rate, to obtain a rice blast area comprehensive change coefficient;

[0054] Step S36: If the rice blast area comprehensive change coefficient is greater than 1, it is judged that the target farmland to be detected is a rice blast spreading farmland;

[0055] Step S37: If the rice blast area comprehensive change coefficient is less than or equal to 1, it is judged that the target farmland to be detected is a rice blast stable farmland;

[0056] Step S38: Perform a spreading trend analysis on the rice blast spreading farmland, and issue a rice blast spreading early warning according to the analysis result.

[0057] Further, the step S32 further includes the following steps:

[0058] The P1 stage farmland marked image is divided into a plurality of F1 type plant squares, the number of F1 type plant squares occupied by the first type of rice plants is obtained to obtain a first plant square number value, and the number of F1 type plant squares occupied by the second type of rice plants is obtained to obtain a second plant square number value;

[0059] The first plant square number value and the second plant square number value are calculated to obtain a P1 image area ratio of the rice blast;

[0060] The P1 image area ratio of the rice blast is calculated, and the specific formula is as follows:

[0061]

[0062] Wherein, Smp1 is the P1 image area ratio of the rice blast, Zgf1 is the first plant square number value, and Zgf2 is the second plant square number value.

[0063] Further, the step S38 further includes the following steps:

[0064] Step S381: In the P1 stage farmland marked image to the Pa stage farmland marked image, the P1 stage farmland marked image and the P2 stage farmland marked image are set as a first image analysis group, the P2 stage farmland marked image and the P3 stage farmland marked image are set as a second image analysis group, and so on, the Pa-1 stage farmland marked image and the Pa stage farmland marked image are set as an e image analysis group;

[0065] Step S382: The first image analysis group is analyzed for the spread of the rice blast, and a first rice blast spatial diffusion value is obtained according to the analysis result;

[0066] Step S383: The second image analysis group to the e image analysis group is obtained respectively, and a second rice blast spatial diffusion value to an e rice blast spatial diffusion value is obtained;

[0067] Step S384: The first rice blast spatial diffusion value to the e rice blast spatial diffusion value is calculated to obtain a farmland rice blast spatial diffusion value;

[0068] Step S385: A farmland rice blast spatial diffusion value preset interval is obtained, if the farmland rice blast spatial diffusion value is in the farmland rice blast spatial diffusion value, it is judged that the rice blast diffusion farmland is in a concentrated diffusion state, and a concentrated diffusion warning is issued, if the farmland rice blast spatial diffusion value is not in the farmland rice blast spatial diffusion value, it is judged that the rice blast diffusion farmland is in a distributed diffusion state, and a distributed diffusion warning is issued.

[0069] Further, the step S382 further includes the following steps:

[0070] Step S3821: Mark the F1 type plant squares in the P1 stage field marking image in the target field plane coordinate system, and perform square fusion on the F1 type plant squares in adjacent state in the target field plane coordinate system to obtain a plurality of F2 type plant squares;

[0071] Step S3822: For the F1 type plant square, the coordinate corresponding to the square center point is marked to obtain a plurality of P1F1 diseased plant coordinates; for the F2 type plant square, the geometric center corresponding to the F2 type plant square is obtained to obtain a plurality of P1F2 diseased plant coordinates.

[0072] Step S3823: Mark the newly added F1 type plant squares of the P2 stage field marking image relative to the P1 stage field marking image in the target field plane coordinate system, and perform square fusion on the newly added F1 type plant squares in adjacent state in the target field plane coordinate system to obtain a plurality of newly added F2 type plant squares.

[0073] Step S3824: For the newly added F1 type plant square, the coordinate corresponding to the square center point is marked to obtain a plurality of P2F1 diseased plant coordinates; for the newly added F2 type plant square, the geometric center corresponding to the newly added F2 type plant square is obtained to obtain a plurality of P2F2 diseased plant coordinates.

[0074] Step S3825: Perform regional diffusion distance analysis on the F1 type plant square, and obtain the F1 square diffusion distance according to the analysis result.

[0075] Step S3826: Perform regional diffusion distance analysis on the F2 type plant square, and obtain the F2 square diffusion distance according to the analysis result.

[0076] The step S3825 further includes the following steps:

[0077] In the target field plane coordinate system, a sample P1F1 diseased plant coordinate is selected from the obtained plurality of P1F1 diseased plant coordinates, the coordinate distance value of the sample P1F1 diseased plant coordinate and each P2F1 diseased plant coordinate is obtained, and the average value of the obtained plurality of coordinate distance values is obtained to obtain the diseased plant diffusion distance corresponding to the sample P1F1 diseased plant coordinate.

[0078] The diseased plant diffusion distance corresponding to each P1F1 diseased plant coordinate is obtained, and the numerical comparison of the obtained plurality of diseased plant diffusion distances is performed, and the diseased plant diffusion distance with the largest value is marked as the F1 square average diffusion distance.

[0079] Further, the step S3826 further includes the following steps:

[0080] In the target farmland plane coordinate system, a sample P1F2 diseased plant coordinate is selected from the obtained plurality of P1F2 diseased plant coordinates, the coordinate distance values of the sample P1F2 diseased plant coordinate and each P2F2 diseased plant coordinate are obtained respectively, and the average value of the obtained plurality of coordinate distance values is obtained to obtain the diseased plant spread distance corresponding to the sample P1F2 diseased plant coordinate;

[0081] The diseased plant spread distance corresponding to each P1F2 diseased plant coordinate is obtained respectively, and the plurality of diseased plant spread distances are compared in value, and the diseased plant spread distance with the maximum value is marked as the second diseased plant spread distance;

[0082] The number of F1 type plant squares in each newly added F2 type plant square is obtained to obtain a plurality of F1 square number values, and the average value of the plurality of F1 square number values is calculated to obtain a square equivalent conversion ratio;

[0083] The F1 square spread distance, the F2 square spread distance and the square equivalent conversion ratio are calculated to obtain a first blight space spread value;

[0084] The first blight space spread value is calculated, and the specific formula is as follows:

[0085]

[0086] Wherein, Ksz1 is the first blight space spread value, Kj1 is the F1 square spread distance, Kj2 is the F2 square spread distance, and Zhb is the square equivalent conversion ratio.

[0087] In summary, due to the adoption of the above technical scheme, the beneficial effects of the present application are:

[0088] 1. The present application analyzes the blight image of the rice leaf in different growth stages and warns the blight of single rice plant, and comprehensively analyzes all the blight images of the rice in the farmland, thereby improving the comprehensiveness of the blight detection result.

[0089] 2. The present application creates a plane coordinate system for the target farmland to be detected, marks the blight plants in the farmland area coordinate system in batches according to the images of rice in different stages, analyzes the stage evolution of the blight plants in the target farmland to be detected according to the marking result, judges the spread range of the blight rice according to the analysis result, and issues a farmland blight warning, which can expand the function of the blight detection method and improve the scene application degree of the blight detection method. BRIEF DESCRIPTION OF DRAWINGS

[0090] In order to facilitate the understanding of those skilled in the art, the present application will be further described below with reference to the drawings.

[0091] In order to facilitate the understanding of those skilled in the art, the present application will be further described below with reference to the drawings.Figure 1 Figure of the embodiment steps of the present application;

[0092] Figure 2 Figure of the target farmland coordinate system of the present application;

[0093] Figure 3 Figure of the P1 stage farmland marking image of the present application. DETAILED DESCRIPTION

[0094] The technical solutions of the present application will be described in detail below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0095] Embodiment one

[0096] Please refer to Figure 1 The present application provides a technical solution: a rice leaf rice blast detection method based on monitoring images, comprising the following specific steps:

[0097] Step S1: obtaining the target farmland to be detected, periodically acquiring images of the target farmland to be detected, obtaining a plurality of rice leaf image sets, and performing leaf image rice blast analysis on each rice plant in each rice leaf image set, and obtaining preliminary analysis data of the leaf image according to the analysis result;

[0098] The step S1 further comprises the following steps:

[0099] Obtaining the rice farmland that needs to be detected for rice blast, and randomly selecting a target farmland to be detected from the obtained plurality of rice farmlands;

[0100] In the process of detecting the target farmland to be detected for rice blast, the time point corresponding to the current time is marked as the cycle end time point to mark a rice disease and pest detection cycle;

[0101] It should be noted that:

[0102] In this application, the period length corresponding to the rice disease and pest detection cycle is generally set to one week. If the planting length of the target farmland to be detected is not full one week, the time point corresponding to the completion of rice planting is marked as the cycle start time point. If the planting length of the target farmland to be detected is not full one week, the time point corresponding to one week before the current time is marked as the cycle start time point;

[0103] A plurality of image acquisition batches are set in the rice disease and pest detection period, and the acquired image acquisition batches are named as P1 image acquisition batch to Pa image acquisition batch in the order of acquisition time;

[0104] It should be noted here that:

[0105] In this application, P here refers to the symbol corresponding to the image acquisition batch, and a here refers to the number value corresponding to the image acquisition batch, and a is an integer greater than 0.

[0106] From the P1 image acquisition batch of the target farmland to be detected, the rice leaf image is acquired to form a P1 rice leaf image set, from the P2 image acquisition batch of the target farmland to be detected, the rice leaf image is acquired to form a P2 rice leaf image set, and so on, from the Pa image acquisition batch of the target farmland to be detected, the rice leaf image is acquired to form a Pa rice leaf image set;

[0107] It should be noted here that:

[0108] In this application, P1 rice leaf image set to Pa rice leaf image set all include the leaf image corresponding to each rice plant in the target farmland to be detected;

[0109] In this application, the image acquisition method and the shooting parameters used by the leaf image are the same, and the shooting parameters here include but are not limited to environmental brightness, shooting focal length and shooting angle.

[0110] The P1 rice leaf image set is analyzed for rice leaf blight, and the plant blight index corresponding to each rice plant is obtained according to the analysis result, and P1 leaf blight analysis data is obtained;

[0111] Specifically as follows:

[0112] The rice plants planted in the target farmland area are acquired, and one of the acquired multiple rice plants is selected as a sample rice plant;

[0113] The rice leaf image corresponding to the sample rice plant is acquired to obtain a plurality of rice leaf images, and a sample rice leaf image is selected from the acquired plurality of rice leaf images;

[0114] The sample rice leaf image is analyzed for leaf blight, and the image blight index corresponding to the sample rice leaf image is obtained according to the analysis result;

[0115] Specifically as follows:

[0116] Marking the rice leaf coverage area in the sample rice leaf image using an image recognition algorithm to obtain an image leaf area;

[0117] Segmenting the image leaf area into a plurality of leaf pixel points, and using an RGB color model to respectively obtain the color R value, color G value and color B value corresponding to each leaf pixel point;

[0118] Obtaining a plurality of rice leaf historical images, and the leaf blight pixel points and healthy leaf pixel points in the rice leaf historical images have been marked;

[0119] It should be noted here that:

[0120] In this application, the leaf blight pixel points referred to herein include blight pixel points corresponding to each leaf blight development stage, and the healthy leaf pixel points referred to herein include leaf pixel points corresponding to each growth stage of the rice leaf.

[0121] Obtaining the color R value interval, color G value interval and color B value interval corresponding to the leaf blight pixel points to obtain a first R value preset interval, a first G value preset interval and a first G value preset interval;

[0122] Obtaining the color R value interval, color G value interval and color B value interval corresponding to the healthy leaf pixel points to obtain a second R value preset interval, a second G value preset interval and a second G value preset interval;

[0123] Obtaining the leaf pixel points whose color R value, color G value and color B value are in the first R value preset interval, the first G value preset interval and the first B value preset interval, respectively, to obtain a plurality of blight area pixel points;

[0124] Respectively obtaining the deviation of the color R value, color G value and color B value corresponding to each blight area pixel point from the second R value preset interval, the second G value preset interval and the second G value preset interval, respectively, to obtain the color R value deviation, color G value deviation and color B value deviation corresponding to each blight area pixel point;

[0125] Performing average number calculation on the color R value deviation, color G value deviation and color B value deviation corresponding to the same blight area pixel point to obtain a plurality of color RGB deviations;

[0126] Counting the number of leaf pixel points in the image leaf area to obtain a first pixel point number value, and counting the number of blight pixel points in the image leaf area to obtain a second pixel point number value;

[0127] Obtaining the image blight index corresponding to the sample rice leaf image by calculating the color RGB deviation, the first pixel point number value and the second pixel point number value.

[0128] The image rice blast index corresponding to the sample rice leaf image is calculated, and the specific formula is as follows:

[0129]

[0130] Wherein, Wbz is the image rice blast index corresponding to the sample rice leaf image, Xsz1 is the first pixel point number value, Xsz2 is the second pixel point number value, Yxs is the color RGB deviation;

[0131] The process of obtaining the image rice blast index corresponding to the sample rice leaf image is repeated, and the image rice blast index corresponding to each rice leaf image is obtained, and the average of the obtained multiple image rice blast indexes is calculated to obtain the plant rice blast index corresponding to the sample rice plant;

[0132] The process of obtaining the plant rice blast index corresponding to the sample rice plant is repeated, and the plant rice blast index corresponding to each rice plant is obtained, and P1 leaf rice blast analysis data is obtained;

[0133] The process of obtaining P1 leaf rice blast analysis data is repeated, and the leaf rice blast analysis data corresponding to P2 rice leaf image set to Pa rice leaf image set is obtained, and P2 leaf rice blast analysis data to Pa leaf rice blast analysis data is obtained;

[0134] P1 leaf rice blast analysis data to Pa leaf rice blast analysis data is defined as leaf image preliminary analysis data;

[0135] Step S2: creating a plane coordinate system for the target farmland to be detected, obtaining a farmland area coordinate system, and marking a batch of rice blast plants in the farmland area coordinate system according to the leaf image preliminary analysis data, obtaining rice blast image marking data;

[0136] In the step S2, the following specific steps are further included:

[0137] A coordinate system is created for the target farmland to be detected, and a target farmland plane coordinate system is obtained;

[0138] Specifically as follows:

[0139] Please refer to Figure 2 , the target farmland is obtained by region plane image acquisition, and the first image feature point corresponding to the target farmland plane image is obtained, and a straight line is made through the first image feature point to obtain a first image feature straight line, and a straight line perpendicular to the first image feature straight line is made through the first image feature point to obtain a second image feature straight line;

[0140] The first image feature point is marked as the coordinate origin, the first image feature straight line is marked as the coordinate x-axis, and the second image feature straight line is marked as the coordinate y-axis, to obtain a target farmland plane coordinate system;

[0141] Obtain leaf image preliminary analysis data, and obtain P1 leaf blight analysis data to Pa leaf blight analysis data according to the leaf image preliminary analysis data;

[0142] According to the P1 leaf blight analysis data, the growth state type of each rice plant is judged, and the judgment result is marked in the target farmland plane coordinate system to obtain a P1 stage farmland marking image;

[0143] Specifically as follows:

[0144] According to the P1 leaf blight analysis data, the corresponding plant blight index of each rice plant in the P1 image collection batch is obtained according to the P1 leaf blight analysis data;

[0145] If the plant blight index is 0, the corresponding rice plant is marked as the first type of rice plant, and if the plant blight index is greater than 0, the corresponding rice plant is marked as the second type of blight rice plant;

[0146] It should be noted here that:

[0147] In this application, the first type of rice plant referred to here is specifically a rice plant that has not been infected with blight, and the second type of rice plant referred to here is specifically a rice plant that has been infected with blight;

[0148] In the target farmland plane image, the first type of rice plant and the second type of rice plant are marked in the target farmland plane coordinate system to obtain a P1 stage farmland marking image;

[0149] Repeat the process of obtaining the P1 stage farmland marking image, and divide the types of the rice plants in the P2 leaf blight analysis data to the Pa leaf blight analysis data and mark the coordinates to obtain the P2 stage farmland marking image to the Pa stage farmland marking image, to obtain blight rice image marking data;

[0150] Step S3: According to the blight rice image marking data, the target farmland to be detected is analyzed for blight plant stage evolution, and a farmland blight warning is issued according to the analysis result;

[0151] In step S3, the following steps are further included:

[0152] Obtain blight rice image marking data, and obtain P1 stage farmland marking image to Pa stage farmland marking image according to the blight rice image marking data;

[0153] The P1 stage farmland marked image is analyzed to obtain a P1 image area ratio of the area of the rice blast area;

[0154] The specific process is as follows:

[0155] The P1 stage farmland marked image is divided into a plurality of F1 type plant squares, the number of the F1 type plant squares occupied by the first type of rice plants is obtained to obtain a first plant square number value, and the number of the F1 type plant squares occupied by the second type of rice plants is obtained to obtain a second plant square number value;

[0156] It should be noted that:

[0157] In the present application, the F1 type plant square contains only one rice plant.

[0158] The first plant square number value and the second plant square number value are calculated to obtain the P1 image area ratio of the area of the rice blast area;

[0159] The P1 image area ratio of the area of the rice blast area is calculated, and the specific formula is as follows:

[0160]

[0161] Wherein, Smp1 is the P1 image area ratio of the area of the rice blast area, Zgf1 is the first plant square number value, and Zgf2 is the second plant square number value.

[0162] The P1 image area ratio of the area of the rice blast area is repeatedly obtained to obtain the image area ratio of the area of the rice blast area corresponding to the P2 stage farmland marked image to the Pa stage farmland marked image, thereby obtaining the P2 image area ratio of the area of the rice blast area to the Pa image area ratio of the area of the rice blast area;

[0163] The difference between the P2 image area ratio of the area of the rice blast area and the P1 image area ratio of the area of the rice blast area is calculated, and the ratio of the obtained difference to the P1 image area ratio of the area of the rice blast area is calculated to obtain a K1 area change rate of the area of the rice blast area, the difference between the P3 image area ratio of the area of the rice blast area and the P2 image area ratio of the area of the rice blast area is calculated, and the ratio of the obtained difference to the P2 image area ratio of the area of the rice blast area is calculated to obtain a K2 area change rate of the area of the rice blast area, and so on, the difference between the Pa image area ratio of the area of the rice blast area and the Pa-1 image area ratio of the area of the rice blast area is calculated, and the ratio of the obtained difference to the Pa-1 image area ratio of the area of the rice blast area is calculated to obtain a Kd area change rate of the area of the rice blast area;

[0164] It should be noted that:

[0165] In the present application, K is a symbol corresponding to the area change rate of the area of the rice blast area, d is a value corresponding to the area change rate of the area of the rice blast area, and k=a-1.

[0166] The K1 rice blast area change rate and the Kd rice blast area change rate are averaged to obtain a rice blast area comprehensive change coefficient;

[0167] If the rice blast area comprehensive change coefficient is greater than 1, the target farmland to be detected is determined as a rice blast spreading farmland;

[0168] If the rice blast area comprehensive change coefficient is less than or equal to 1, the target farmland to be detected is determined as a rice blast stable farmland;

[0169] It should be noted here that:

[0170] In the present application, the rice blast stable farmland referred to here means a farmland in which rice blast has occurred but the disease condition has been controlled and no longer continues to spread, and the rice blast spreading farmland means a farmland in which rice blast has occurred and the disease condition is spreading to the surrounding healthy area;

[0171] The rice blast spreading farmland is analyzed for a spreading trend, and a rice blast spreading early warning is issued according to the analysis result;

[0172] Specifically as follows:

[0173] In the P1 stage farmland marked image to the Pa stage farmland marked image, the P1 stage farmland marked image and the P2 stage farmland marked image are set as a first image analysis group, the P2 stage farmland marked image and the P3 stage farmland marked image are set as a second image analysis group, and so on, the Pa-1 stage farmland marked image and the Pa stage farmland marked image are set as an e-th image analysis group;

[0174] It should be noted here that:

[0175] In the present application, e here refers to a quantity value corresponding to the image analysis group, and e is an integer greater than 0;

[0176] The first image analysis group is analyzed for a rice blast plant spreading, and a first rice blast spatial spreading value is obtained according to the analysis result;

[0177] Specifically as follows:

[0178] Please refer to Figure 3 The F1 type plant squares in the P1 stage farmland marked image are marked in the target farmland plane coordinate system, and the F1 type plant squares in the target farmland plane coordinate system in an adjacent state are fused to obtain a plurality of F2 type plant squares;

[0179] For the F1 type plant square, the coordinates corresponding to the center point of the square are marked to obtain a plurality of P1F1 diseased plant coordinates; for the F2 type plant square, the coordinates of the geometric center corresponding to the F2 type plant square are obtained to obtain a plurality of P1F2 diseased plant coordinates;

[0180] The newly added F1 type plant square of the P2 stage agricultural field marking image relative to the P1 stage agricultural field marking image is marked in the target agricultural field plane coordinate system, and the newly added F1 type plant squares in the adjacent state in the target agricultural field plane coordinate system are fused to obtain a plurality of newly added F2 type plant squares;

[0181] For the newly added F1 type plant square, the coordinates corresponding to the center point of the square are marked to obtain a plurality of P2F1 diseased plant coordinates; for the newly added F2 type plant square, the coordinates of the geometric center corresponding to the newly added F2 type plant square are obtained to obtain a plurality of P2F2 diseased plant coordinates;

[0182] The F1 type plant square is subjected to regional diffusion distance analysis, and the F1 square diffusion distance is obtained according to the analysis result;

[0183] Specifically as follows:

[0184] In the target agricultural field plane coordinate system, a sample P1F1 diseased plant coordinate is selected from the obtained plurality of P1F1 diseased plant coordinates, the coordinate distance values of the sample P1F1 diseased plant coordinate and each P2F1 diseased plant coordinate are obtained, and the average value of the obtained plurality of coordinate distance values is taken to obtain the diseased plant diffusion distance corresponding to the sample P1F1 diseased plant coordinate;

[0185] The process of obtaining the diseased plant diffusion distance corresponding to the sample P1F1 diseased plant coordinate is repeated, the diseased plant diffusion distance corresponding to each P1F1 diseased plant coordinate is obtained, and the plurality of diseased plant diffusion distances are subjected to numerical comparison, and the diseased plant diffusion distance with the largest value is marked as the F1 square average diffusion distance;

[0186] The F2 type plant square is subjected to regional diffusion distance analysis, and the F2 square diffusion distance is obtained according to the analysis result;

[0187] Specifically as follows:

[0188] In the target agricultural field plane coordinate system, a sample P1F2 diseased plant coordinate is selected from the obtained plurality of P1F2 diseased plant coordinates, the coordinate distance values of the sample P1F2 diseased plant coordinate and each P2F2 diseased plant coordinate are obtained, and the average value of the obtained plurality of coordinate distance values is taken to obtain the diseased plant diffusion distance corresponding to the sample P1F2 diseased plant coordinate;

[0189] Repeating the process of obtaining the strain spread distance corresponding to the P1F2 strain coordinates of the sample, obtaining the strain spread distance corresponding to each P1F2 strain coordinate, and comparing the obtained multiple strain spread distances, the maximum strain spread distance is marked as the second strain spread distance;

[0190] Obtaining the number of F1 type plant squares in each newly added F2 type plant square, obtaining multiple F1 square number values, and calculating the average value of the obtained multiple F1 square number values to obtain the square equivalent conversion ratio;

[0191] The F1 square spread distance, F2 square spread distance and square equivalent conversion ratio are calculated to obtain the first blight space diffusion value;

[0192] The first blight space diffusion value is calculated, and the specific formula is as follows:

[0193]

[0194] Wherein, Ksz1 is the first blight space diffusion value, Kj1 is the F1 square spread distance, Kj2 is the F2 square spread distance, and Zhb is the square equivalent conversion ratio;

[0195] Repeating the process of obtaining the first blight space diffusion value, obtaining the blight space diffusion value corresponding to the second image analysis group to the e image analysis group, obtaining the second blight space diffusion value to the e blight space diffusion value;

[0196] And the average value of the first blight space diffusion value to the e blight space diffusion value is calculated to obtain the farmland blight space diffusion value;

[0197] Obtaining the farmland blight space diffusion value preset interval, if the farmland blight space diffusion value is in the farmland blight space diffusion value, it is judged that the rice blight diffusion farmland is in the concentrated diffusion state, and the concentrated diffusion warning is issued, if the farmland blight space diffusion value is not in the farmland blight space diffusion value, it is judged that the rice blight diffusion farmland is in the distributed diffusion state, and the distributed diffusion warning is issued;

[0198] It should be noted here that:

[0199] Obtaining a plurality of target farmland in the concentrated diffusion state of rice planting monitoring period, obtaining the farmland blight space diffusion value corresponding to each rice planting monitoring period, marking the minimum farmland blight space diffusion value as the first space diffusion reference value, marking the maximum farmland blight space diffusion value as the second space diffusion reference value, and marking the numerical interval composed of the first space diffusion reference value and the second space diffusion reference value as the farmland blight space diffusion value preset interval.

[0200] In the present application, if the corresponding calculation formula appears, the above calculation formula is all de-dimensioned to calculate the numerical value, and the weight coefficient, the proportional coefficient and other coefficients existing in the formula are set to a result value quantified by each parameter. The size of the weight coefficient and the proportional coefficient only needs to not affect the proportional relationship between the parameter and the result value.

[0201] The preferred embodiments of the application disclosed above are only used to help explain the application. The preferred embodiments do not describe all the details and limit the application to the specific embodiments. Obviously, according to the content of the specification, many modifications and changes can be made. The specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the application, so that those skilled in the art can well understand and utilize the application. The application is limited by the claims and their entire scope and equivalents.

Claims

1. A method for detecting rice blast on rice leaves based on monitoring images, characterized in that: include: Step S1: acquiring a target farmland to be inspected, periodically acquiring images of the target farmland to be inspected, obtaining a plurality of rice leaf image sets, performing rice blast disease analysis on the leaf images of the rice plants in each rice leaf image set, and obtaining preliminary leaf image analysis data based on the analysis results; Step S2: creating a plane coordinate system for the target farmland to be inspected to obtain a farmland region coordinate system, marking batches of blast-infected plants in the farmland region coordinate system based on preliminary leaf image analysis data to obtain blast-infected rice image marking data; Step S3: Analyze the stage-by-stage evolution of diseased plants in the target farmland to be detected based on the diseased rice image marking data, and issue a farmland disease warning based on the analysis results.

2. The method for detecting rice blast on rice leaves based on monitoring images according to claim 1, characterized in that: The step S1 further includes the following steps: Step S11: obtaining rice farmlands that need to be tested for rice blast disease, and arbitrarily selecting a target farmland to be tested from the obtained multiple rice farmlands; Step S12: During the process of performing rice blast detection on the target farmland to be detected, a rice pest and disease detection cycle is marked by taking the time point corresponding to the current moment as the end time point of the cycle; Step S13: setting image acquisition batches P1 to Pa within the rice pest and disease detection cycle; Step S14: obtaining rice leaf images from the P1 image acquisition batch of the target farmland to be inspected to form a P1 rice leaf image set. Similarly, obtaining rice leaf images from the Pa image acquisition batch of the target farmland to be inspected to form a Pa rice leaf image set. Step S15: performing rice leaf blast disease analysis on the P1 rice leaf image set, obtaining a plant blast disease index corresponding to each rice plant based on the analysis results, and obtaining P1 leaf blast disease analysis data; Step S16: respectively acquiring the leaf blast disease analysis data corresponding to the rice leaf image set P2 to the rice leaf image set Pa, and obtaining the leaf blast disease analysis data P2 to the leaf blast disease analysis data Pa; Step S17: defining the P1 leaf blast analysis data to the Pa leaf blast analysis data as preliminary leaf image analysis data.

3. The method for detecting rice blast on rice leaves based on monitoring images according to claim 2, characterized in that: The step S15 further includes the following steps: Step S151: obtaining rice plants grown in the target farmland area to be inspected, and randomly selecting one of the obtained multiple rice plants as a sample rice plant; Step S152: acquiring rice leaf images corresponding to the sample rice plants to obtain a plurality of rice leaf images, and selecting a sample rice leaf image from the plurality of acquired rice leaf images; Step S153: performing leaf blast analysis on the sample rice leaf image, and obtaining an image blast index corresponding to the sample rice leaf image according to the analysis result; Step S154: obtaining the image blast index corresponding to each rice leaf image, and averaging the obtained multiple image blast indices to obtain the plant blast index corresponding to the sample rice plant; Step S155: Obtain the plant blast index corresponding to each rice plant to obtain P1 leaf blast analysis data.

4. The method for detecting rice blast on rice leaves based on monitoring images according to claim 3, characterized in that: The step S153 further includes the following steps: Using an image recognition algorithm to mark the rice leaf coverage area in the sample rice leaf image to obtain the image leaf area; The leaf area of ​​the image is divided into several leaf pixels, and the RGB color model is used to calculate the color R value, color G value, and color B value corresponding to each leaf pixel; Acquire multiple historical images of rice leaves, wherein both leaf blast pixel points and healthy leaf pixel points in the historical images of rice leaves are marked; Obtain the color R value interval, color G value interval, and color B value interval corresponding to the leaf blight pixel point, and obtain a first R value preset interval, a first G value preset interval, and a first B value preset interval; Obtain the color R value interval, color G value interval, and color B value interval corresponding to the healthy leaf pixel point, and obtain a second R value preset interval, a second G value preset interval, and a second B value preset interval; Acquire leaf pixel points whose color R value, color G value, and color B value are respectively within a first R value preset interval, a first G value preset interval, and a first B value preset interval to obtain a plurality of plague area pixel points; Obtain the deviations of the color R value, color G value, and color B value corresponding to each pixel point in the plague area from the second R value preset interval, the second G value preset interval, and the second G value preset interval, respectively, to obtain the color R value deviation, color G value deviation, and color B value deviation corresponding to each pixel point in the plague area; The color R value deviation, color G value deviation, and color B value deviation corresponding to the pixel point in the same plague area are averaged to obtain multiple color RGB deviations; Counting the number of leaf pixels in the leaf area of ​​the image to obtain a first pixel number value, and counting the number of blast pixels in the leaf area of ​​the graphic to obtain a second pixel number value; The color RGB deviation, the first pixel number value, and the second pixel number value are calculated to obtain the image blast index corresponding to the sample rice leaf image.

5. The method for detecting rice blast on rice leaves based on monitoring images according to claim 1, characterized in that: The step S2 further includes the following specific steps: Step S21: creating a coordinate system for the target farmland to be detected to obtain a plane coordinate system of the target farmland; Step S22: obtaining preliminary analysis data of the leaf image, and obtaining P1 leaf blast analysis data to Pa leaf blast analysis data respectively according to the preliminary analysis data of the leaf image; Step S23: judging the growth status type of each rice plant based on the P1 leaf blast analysis data, and marking the judgment results in the target farmland plane coordinate system to obtain a P1 stage farmland marked image; Step S24: performing type classification and coordinate marking on the rice plants in the P2 leaf blast analysis data to the Pa leaf blast analysis data, respectively, to obtain the P2 stage farmland marked image to the Pa stage farmland marked image, and obtain the blast rice image marked data; The step S23 further includes the following specific steps: According to the P1 leaf blast analysis data, the plant blast index corresponding to each rice plant in the P1 image acquisition batch is obtained respectively; If the plant blast index is 0, the corresponding rice plant is marked as a first type rice plant; if the plant blast index is greater than 0, the corresponding rice plant is marked as a second type blast rice plant; In the target farmland plane image, the first type of rice plants and the second type of rice plants are marked in the target farmland plane coordinate system to obtain a P1 stage farmland marked image.

6. The method for detecting rice blast on rice leaves based on monitoring images according to claim 1, characterized in that: The step S3 further includes the following steps: Step S31: obtaining the image marking data of the diseased rice, and obtaining the staged farmland marking images P1 to Pa according to the image marking data of the diseased rice; Step S32: performing a diseased area analysis on the P1 stage marked farmland image, and obtaining the diseased area ratio of the P1 image according to the analysis result; Step S33: respectively obtaining the image blight area ratios corresponding to the farmland marked image at the P2 stage to the farmland marked image at the Pa stage, and obtaining the image blight area ratio of the P2 image to the image blight area ratio of the Pa image; Step S34: Calculate the difference between the P2 image's plague area ratio and the P1 image's plague area ratio, and calculate the ratio of the obtained difference to the P1 image's plague area ratio to obtain the K1 plague area change rate. Similarly, calculate the difference between the Pa image's plague area ratio and the Pa-1 image's plague area ratio, and calculate the ratio of the obtained difference to the Pa-1 image's plague area ratio to obtain the Kd plague area change rate. Step S35: Calculate the average of the K1 plague area change rate and the Kd plague area change rate to obtain the plague area comprehensive change coefficient; Step S36: If the comprehensive coefficient of variation of the diseased area is greater than 1, the target farmland to be inspected is determined to be a farmland with rice blast spreading; Step S37: If the comprehensive coefficient of variation of the blast area is less than or equal to 1, the target farmland to be inspected is determined to be a stable farmland for rice blast; Step S38: Analyze the spread trend of rice blast in the farmland and issue a blast spread warning based on the analysis results.

7. The method for detecting rice blast on rice leaves based on monitoring images according to claim 6, characterized in that: The step S32 further includes the following steps: Divide the P1 stage farmland marked image into several F1 type plant grids, obtain the number of F1 type plant grids occupied by the first type rice plants, and obtain the number of F1 type plant grids occupied by the second type rice plants to obtain the number of second plant grids; The first plant grid quantity value and the second plant grid quantity value are calculated to obtain the diseased area ratio of the P1 image.

8. The method for detecting rice blast on rice leaves based on monitoring images according to claim 6, characterized in that: The step S38 further includes the following steps: Step S381: among the farmland labeled images from stage P1 to stage Pa, the farmland labeled images from stage P1 and stage P2 are set as the first image analysis group. Similarly, the farmland labeled images from stage Pa-1 and stage Pa are set as the e-th image analysis group. Step S382: performing a diseased plant diffusion analysis on the first image analysis group, and obtaining a first disease spatial diffusion value according to the analysis result; Step S383: respectively obtaining the pest spatial diffusion values ​​corresponding to the second image analysis group to the e-th image analysis group, and obtaining the second pest spatial diffusion value to the e-th pest spatial diffusion value; Step S384: Calculate the average value of the spatial diffusion value of the first disease to the spatial diffusion value of the e-th disease to obtain the spatial diffusion value of the farmland disease; Step S385: Obtain a preset interval of the spatial diffusion value of farmland blast. If the spatial diffusion value of farmland blast is within the spatial diffusion value of farmland blast, it is determined that the farmland where rice blast spreads is in a concentrated diffusion state, and a concentrated diffusion warning is issued. If the spatial diffusion value of farmland blast is not within the spatial diffusion value of farmland blast, it is determined that the farmland where rice blast spreads is in a distributed diffusion state, and a distributed diffusion warning is issued.

Images