An agricultural machinery control system based on Beidou navigation and positioning technology

Abstract

The invention discloses an agricultural machinery control system based on Beidou navigation and positioning technology, relates to the field of positioning and navigation technology, comprises a control center, wherein the control center is connected with a path data acquisition module, an irrigation path generation module, a growth image acquisition module and a harvesting path generation module; the path data of a farmland is obtained by a sowing agricultural machinery; the irrigation radius of the irrigation agricultural machinery is obtained, and an irrigation path is generated according to the irrigation radius and the path data; an irrigation cycle is set, and the irrigation agricultural machinery obtains the growth image of crops through the irrigation path according to the irrigation cycle; whether there are diseased plants in the crops is determined according to the growth image, and if there are diseased plants, the harvesting position is obtained; the plant distance is obtained according to the growth image, and a temporary path node is generated according to the plant distance; the geographical location of the harvesting agricultural machinery is obtained by Beidou positioning, and recorded as a departure place; the harvesting position is recorded as a destination; and a harvesting path is generated according to the departure place, the destination, the temporary path node and the path data.

Description

Technical Field

[0001] The present invention relates to the field of positioning and navigation technology, and in particular to an agricultural machinery management and control system based on Beidou navigation and positioning technology. Background Art

[0002] The application history of Beidou in agricultural machinery management can be traced back to around 2012, when China began to promote the Beidou satellite navigation system in the agricultural field. With the continuous improvement and popularization of the Beidou system, its application in agricultural machinery management has gradually increased, mainly in positioning navigation, intelligent scheduling, operation monitoring, intelligent pesticide and fertilization, and data management and analysis.

[0003] When a crop becomes diseased, it is very likely to infect surrounding crops and cause economic losses. How to coordinate agricultural machinery to efficiently eradicate diseased crops to minimize the economic losses caused by infected crops is an issue we need to consider at present. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide an agricultural machinery management and control system based on Beidou navigation and positioning technology.

[0005] The object of the present invention can be achieved by the following technical solutions: an agricultural machinery control system based on Beidou navigation and positioning technology, comprising a control center, wherein the control center is communicatively connected with a path data acquisition module, an irrigation path generation module, a growth image acquisition module and a harvesting path generation module;

[0006] The path data acquisition module is used to set a Beidou short message device on the sowing agricultural machine to obtain the path data of the farmland through the sowing agricultural machine, and the path data includes path nodes and basic paths;

[0007] The irrigation path generation module is used to obtain the irrigation radius of the irrigation agricultural machinery and generate the irrigation path according to the irrigation radius and path data;

[0008] The growth image acquisition module is used to set the irrigation cycle. The irrigation agricultural machine irrigates the crops through the irrigation path according to the irrigation cycle and obtains the growth image of the crops; it is judged whether there are diseased plants in the crops according to the growth image. If there are diseased plants, the harvesting position is obtained; if there are no diseased plants, no operation is required;

[0009] The harvesting path generation module is used to obtain the plant distance according to the growth image, and generate temporary path nodes according to the plant distance; obtain the geographical location of the harvesting agricultural machinery through Beidou positioning, and record it as the departure place; obtain the harvesting position, and record it as the destination; generate a harvesting path according to the departure place, destination, temporary path nodes and path data, and Beidou navigation guides the harvesting agricultural machinery to harvest the diseased plants according to the harvesting path.

[0010] Furthermore, a Beidou short message device is set on the sowing agricultural machinery, and the process of obtaining the path data of the farmland through the sowing agricultural machinery includes:

[0011] A Beidou short message device is set on the seeding agricultural machinery to obtain the driving data of the seeding agricultural machinery in real time through Beidou positioning. The driving data includes the driving position and driving time. The driving position is composed of longitude and latitude. The driving position and driving time are in a one-to-one correspondence.

[0012] The driving data is packaged and sent to the Beidou short message device, and the driving data is transmitted to the path data collection module through the Beidou short message device;

[0013] The path data acquisition module generates a driving path according to the driving position and driving time in the driving data, locates the driving position in order from morning to night according to its corresponding driving time, and connects the locations in sequence to generate a driving path;

[0014] The driving positions in the three driving data are obtained in chronological order, which are recorded as D1 (LON1, LAT1), D2 (LON2, LAT2) and D3 (LON3, LAT3), where LON is longitude and LAT is latitude;

[0015] Set the error threshold LLE;

[0016] Get standard longitude and standard latitude

[0017] like Then the driving position D2 is marked as a path node; otherwise, no operation is required;

[0018] If there is a driving path intersection in the driving path, the intersection position of the driving path is obtained, and the intersection position is marked as a path node;

[0019] If there is a driving path between the path nodes, it is marked as a basic path.

[0020] Furthermore, the process of obtaining the irrigation radius of the irrigation agricultural machinery and generating the irrigation path according to the irrigation radius and the path data includes:

[0021] Acquire the irrigation radius of the irrigation agricultural machinery, wherein the irrigation radius includes the left irrigation radius and the right irrigation radius; simulate the basic driving path of the irrigation agricultural machinery, and acquire the unilateral irrigation range of the basic path according to the irrigation radius of the irrigation agricultural machinery, wherein the unilateral irrigation range includes the left unilateral irrigation range and the right unilateral irrigation range;

[0022] The unilateral irrigation overlap rate of each basic path is obtained according to the unilateral irrigation range of each basic path. The unilateral irrigation overlap rate includes the left unilateral irrigation overlap rate and the right unilateral irrigation overlap rate. The unilateral irrigation overlap rate is the ratio of the overlapping range of the unilateral irrigation range of the basic path and the unilateral irrigation range of other basic paths to the unilateral irrigation range of the basic path, which is expressed as follows:

[0023]

[0024] Set the overlap threshold OLP;

[0025] If the left unilateral irrigation overlap rate and the right unilateral irrigation overlap rate of basic path A are both greater than the overlap threshold OLP, no operation is required, that is, basic path A is not used as an irrigation path; otherwise, basic path A is marked as an irrigation path.

[0026] Furthermore, the growth image of the crops is obtained, and whether there are diseased plants in the crops is determined according to the growth image. If there are diseased plants, the process of obtaining the harvesting position includes:

[0027] Cameras and processors are arranged on the left and right sides of the irrigation machine;

[0028] Set the image acquisition cycle IAC;

[0029] The camera is used to obtain growth images of crops according to an image acquisition cycle IAC;

[0030] The processor is used to generate a unique image number based on the growth image, and obtain the current driving position through Beidou positioning, which is recorded as an image landmark; the growth image, the image number and the image landmark are in a one-to-one correspondence with each other; at the same time, it is used to detect in real time whether there are diseased plants in the crops in the growth image. If there are diseased plants, the image landmark corresponding to the growth image is obtained, and the image landmark is marked as the harvesting position.

[0031] Furthermore, the process of detecting whether there are diseased plants in the crops in the growth image includes:

[0032] Step a: Model each stage of the crop pathological process through a convolutional neural network, obtain several pathological models, and generate a pathological model set;

[0033] Repeat step a, model the pathological process of different diseases of crops through convolutional neural network, obtain several pathological model sets, and generate plant abnormality sets;

[0034] Through convolutional neural networks, various stages of crop growth are modeled, several stem and leaf models are obtained, and a plant morphology set is generated;

[0035] Recognizing the growth image according to the plant morphology set, identifying the region of the crop in the growth image, and marking the region as a plant region to generate a plant region set;

[0036] Establish a two-dimensional rectangular coordinate system A, and map the growth image into the two-dimensional rectangular coordinate system A;

[0037] Get the coordinate with the smallest y value in the plant area and record it as the bottom line coordinate;

[0038] Get the minimum y value of all plant areas in the plant area set to generate a level set; get the median in the level set and mark it as the level value;

[0039] Set y = horizontal value (x∈R) as the horizontal line of the plant;

[0040] Get the distance between the bottom line coordinate and the horizontal line of the plant, and record it as the bottom line distance;

[0041] Set distance threshold;

[0042] If the bottom line distance > the distance threshold, the plant area is deleted from the plant area set;

[0043] If the bottom line distance is ≤ the distance threshold, no action is required;

[0044] identifying plant regions in the plant region set according to the plant anomaly set, and identifying diseased regions in the plant region indicating crop disease;

[0045] If there is a diseased area in the plant area, the plant area is marked as a diseased plant; if there is no diseased area in the plant area, no operation is required.

[0046] Furthermore, the process of obtaining the plant distance according to the growth image and generating a temporary path node according to the plant distance includes:

[0047] Set the bottom threshold BT;

[0048] Establish a two-dimensional rectangular coordinate system B, and map the growth image into the two-dimensional rectangular coordinate system B;

[0049] Get the minimum y value in the plant area and mark it as the bottom y value;

[0050] Obtain the top y value according to the bottom y value, wherein the top y value=bottom y value+bottom threshold BT;

[0051] Generate a bottom region in a two-dimensional rectangular coordinate system B according to the bottom y value and the top y value: bottom y value ≤ y ≤ top y value (x∈R); obtain an area where the plant area overlaps with the bottom region, and mark the area as the main stem area of ​​the plant;

[0052] Acquire stem images and leaf images at various stages of crop growth to generate a number of stem-leaf image pairs [stem image, leaf image]; the stem-leaf image pair is a stem image and a leaf image at the same stage, and the stem image only contains the stem of the crop, and the leaf image only contains the leaf of the crop;

[0053] Step b1: Obtain the pixel value corresponding to each pixel point in the stem image, generate a stem pixel set, compare the pixel values ​​in the stem pixel set, obtain the maximum value and the minimum value, mark the maximum value as the stem peak value, and mark the minimum value as the stem valley value; similarly, obtain the leaf pixel set, leaf peak value and leaf valley value of the leaf image;

[0054] Step b2: Generate a stem color value range [stem valley value, stem peak value] according to the stem peak value and the stem valley value, and generate a leaf color value range [leaf valley value, leaf peak value] according to the leaf peak value and the leaf valley value; if there is an overlapping area in the stem color value range and the leaf color value range in terms of value, then remove the overlapping area in the stem color value range and the leaf color value range,

[0055] Step b3: Obtain the mode of the pixel values ​​in the stem pixel set that belong to the stem color value range, recorded as the stem color mode; similarly, obtain the leaf color mode; obtain

[0056] Step b4: generating stem and leaf characteristic factors according to the stem color range, the leaf color range and the stem and leaf contrast value;

[0057] Repeat steps b1 to b4 to obtain the stem-leaf feature factors of the stem-leaf image pairs at each stage and generate a stem-leaf feature set;

[0058] Set the light error threshold RET, and the light error threshold RET is a percentage;

[0059] Obtain the pixel value of each pixel point in the main stem area of ​​the plant to generate a pixel set in the bottom area;

[0060] Step c1: Filter the pixel values ​​in the bottom area pixel set, filter out the pixel values ​​whose pixel values ​​∈ the stem color value range, and generate the bottom stem pixel set according to the pixel values, and then filter out the pixel values ​​whose pixel values ​​∈ the leaf color value range, and generate the bottom leaf pixel set according to the pixel values;

[0061] Step c2: Obtain the majority of the bottom stem pixel set, recorded as the bottom stem majority; obtain the majority of the bottom leaf pixel set, recorded as the bottom leaf majority; generate the bottom stem majority and the bottom leaf majority according to the bottom stem majority and the bottom leaf majority. If [stem-leaf contrast value × (1-RET)] ≤ bottom stem-leaf contrast value ≤ [stem-leaf contrast value × (1+RET)], then the bottom stem mode is marked as the stem color value; otherwise, no operation is required;

[0062] According to the stem-leaf feature set, steps c1 and c2 are repeated. If a stem color value is obtained, it is marked as a characteristic color value; if several stem color values ​​are obtained, a stem color value is randomly selected and marked as a characteristic color value; if no stem color value is obtained, a warning message is generated to inform the user that the recognition fails;

[0063] Step d1: obtaining the coordinates corresponding to the characteristic color value of the pixel points in the main stem area of ​​the plant, marking them as stem coordinates, and generating a stem coordinate set;

[0064] Step d2: Obtain the coordinate with the largest x value in the stem coordinate set, marked as the stem X peak coordinate; obtain the coordinate with the smallest x value in the stem coordinate set, marked as the stem X valley coordinate; obtain the coordinate with the largest y value in the stem coordinate set, marked as the stem Y peak coordinate; obtain the coordinate with the smallest y value in the stem coordinate set, marked as the stem Y valley coordinate; generate the stem center coordinate according to the stem X peak coordinate, the stem X valley coordinate, the stem Y peak coordinate and the stem Y valley coordinate, and the stem center coordinate is expressed as follows:

[0065]

[0066] Repeat steps d1 and d2 to obtain the stem center coordinates of each plant region in the plant region set to generate a stem center coordinate set;

[0067] Obtain the adjacent stem center coordinates according to the x value, and generate a stem center coordinate pair (stem center coordinate A, stem center coordinate B);

[0068] Get stem distance according to stem center coordinate pair = |x 茎中心坐标A -x 茎中心坐标B |;

[0069] Set the road width threshold RWT;

[0070] If the stem distance < the road width threshold RWT, no action is required;

[0071] If the stem distance ≥ the road width threshold RWT, obtain the temporary path longitudinal line (y∈R); obtain the growth image ratio and the image landmark corresponding to the growth image, and generate a temporary path node according to the temporary path vertical line, the growth image ratio and the image landmark, wherein the temporary path node is composed of longitude and latitude.

[0072] Furthermore, the process of generating a harvesting path according to the departure place, the destination, the temporary path node and the path data includes:

[0073] Establish a geographic coordinate system and map the path nodes, basic paths and temporary path nodes to the geographic coordinate system;

[0074] Get the growth image of the generated temporary path nodes and mark it as the path image;

[0075] Obtain an image number corresponding to the path image, obtain a processor that generates the image number, and query an instrument orientation of the processor, wherein the instrument orientation includes a left side of the irrigation agricultural machine and a right side of the irrigation agricultural machine;

[0076] If the instrument is on the left side of the irrigation machine, the temporary path node is taken as the origin, perpendicular to the basic path where the temporary path node is located, and extended to the adjacent path node or basic path to generate a temporary path; if the instrument is on the right side of the irrigation machine, the temporary path node is taken as the origin, perpendicular to the basic path where the temporary path node is located, and extended to the adjacent path node or basic path to generate a temporary path;

[0077] The shortest path from the departure point to the destination is generated according to the departure point, the destination, the temporary path node, the temporary path, the path node and the basic path, and the shortest path is marked as a harvesting path.

[0078] Compared with the prior art, the present invention has the following beneficial effects:

[0079] 1. The present invention sets an irrigation cycle so that the irrigation agricultural machine periodically obtains growth images according to the irrigation cycle, then obtains the plant distance according to the growth image, and generates temporary path nodes according to the plant distance, that is, the temporary path nodes will be updated according to the cycle; by periodically updating the temporary path nodes, the driving path is made more flexible and effective, so that the diseased plants can be harvested quickly and effectively in the future;

[0080] 2. The present invention obtains the main stem area of ​​the plant area, obtains the stem center coordinates in the main stem area, and then obtains the stem center coordinates in each plant area. The distance between the plants is calculated by the stem center coordinates, that is, the distance between the plants is calculated according to the stem close to the root part, thereby avoiding the risk of crushing crops when generating temporary path nodes to a certain extent;

[0081] 3. The present invention eliminates crops beside other basic paths through the horizontal line of plants, thereby avoiding interference of crops beside other basic paths in calculating the distance between plants beside the basic path to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0083] like Figure 1 As shown, an agricultural machinery management and control system based on Beidou navigation and positioning technology includes a control center, and the control center is connected to a path data acquisition module, an irrigation path generation module, a growth image acquisition module and a harvesting path generation module;

[0084] The path data acquisition module is used to set a Beidou short message device on the sowing agricultural machine to obtain the path data of the farmland through the sowing agricultural machine, and the path data includes path nodes and basic paths;

[0085] The irrigation path generation module is used to obtain the irrigation radius of the irrigation agricultural machinery and generate the irrigation path according to the irrigation radius and path data;

[0086] The growth image acquisition module is used to set the irrigation cycle. The irrigation agricultural machine irrigates the crops through the irrigation path according to the irrigation cycle and obtains the growth image of the crops; it is judged whether there are diseased plants in the crops according to the growth image. If there are diseased plants, the harvesting position is obtained; if there are no diseased plants, no operation is required;

[0087] The harvesting path generation module is used to obtain the plant distance according to the growth image, and generate a temporary path node according to the plant distance; obtain the geographical location of the harvesting agricultural machinery through Beidou positioning, and record it as the departure place; obtain the harvesting position, and record it as the destination; generate a harvesting path according to the departure place, the destination, the temporary path node and the path data, and the Beidou navigation guides the harvesting agricultural machinery to harvest the diseased plants according to the harvesting path;

[0088] Specifically, the process of setting up Beidou short message equipment on a sowing agricultural machine and obtaining the path data of the farmland through the sowing agricultural machine includes:

[0089] A Beidou short message device is installed on the sowing agricultural machinery to obtain the driving data of the sowing agricultural machinery in real time through Beidou positioning, wherein the driving data includes the driving position and driving time, and the driving position is composed of longitude and latitude;

[0090] It should be noted that the driving position and driving time are in a one-to-one correspondence; the driving data is packaged and sent to the Beidou short message device, and the driving data is transmitted to the path data collection module through the Beidou short message device;

[0091] The path data acquisition module generates a driving path according to the driving position and driving time in the driving data, locates the driving position in order from morning to night according to its corresponding driving time, and connects the locations in sequence to generate a driving path;

[0092] The driving positions in the three driving data are obtained in chronological order, which are recorded as D1 (LON1, LAT1), D2 (LON2, LAT2) and D3 (LON3, LAT3), where LON is longitude and LAT is latitude;

[0093] Set the error threshold LLE;

[0094] Get standard longitude and standard latitude

[0095] like Then the driving position D2 is marked as a path node; otherwise, no operation is required;

[0096] If there is a driving path intersection in the driving path, the intersection position of the driving path is obtained, and the intersection position is marked as a path node;

[0097] If there is a driving path between the path nodes, it is marked as a basic path;

[0098] Specifically, the process of obtaining the irrigation radius of the irrigation agricultural machinery and generating the irrigation path according to the irrigation radius and path data includes:

[0099] Acquire the irrigation radius of the irrigation agricultural machine, where the irrigation radius includes a left irrigation radius and a right irrigation radius, wherein the left irrigation radius is the irrigation distance on the left side of the irrigation agricultural machine, and the right irrigation radius is the irrigation distance on the right side of the irrigation agricultural machine;

[0100] Simulate the basic travel path of the irrigation agricultural machinery, and obtain the unilateral irrigation range of the basic path according to the irrigation radius of the irrigation agricultural machinery, wherein the unilateral irrigation range is composed of a left unilateral irrigation range and a right unilateral irrigation range, wherein the left unilateral irrigation range is the irrigation range generated according to the left irrigation radius when the irrigation agricultural machinery travels the basic path, and the right unilateral irrigation range is the irrigation range generated according to the right irrigation radius when the irrigation agricultural machinery travels the basic path;

[0101] The unilateral irrigation overlap rate of each basic path is obtained according to the unilateral irrigation range of each basic path. The unilateral irrigation overlap rate includes the left unilateral irrigation overlap rate and the right unilateral irrigation overlap rate. The unilateral irrigation overlap rate is the ratio of the overlapping range of the unilateral irrigation range of the basic path and the unilateral irrigation range of other basic paths to the unilateral irrigation range of the basic path, which is expressed as follows:

[0102]

[0103] Set the overlap threshold OLP;

[0104] If the left unilateral irrigation overlap rate and the right unilateral irrigation overlap rate of the basic path A are both greater than the overlap threshold OLP, no operation is required, that is, the basic path A is not used as an irrigation path; otherwise, the basic path A is marked as an irrigation path;

[0105] Specifically, the process of obtaining the growth image of the crops and determining whether there are diseased plants in the crops according to the growth image, and if there are diseased plants, obtaining the harvesting position includes:

[0106] The cameras and processors are arranged on the left and right sides of the irrigation machine, that is, the cameras and processors are arranged on the side facing the crops;

[0107] Set the image acquisition cycle IAC;

[0108] The camera is used to obtain growth images of crops according to an image acquisition cycle IAC;

[0109] The processor is used to generate a unique image number according to the growth image, and obtain the current driving position through Beidou positioning, which is recorded as an image landmark; the growth image, the image number and the image landmark are in a one-to-one correspondence with each other; at the same time, it is used to detect in real time whether there are diseased plants in the crops in the growth image, and if there are diseased plants, the image landmark corresponding to the growth image is obtained, and the image landmark is marked as the harvesting position;

[0110] The process of detecting whether there are diseased plants in the crops in the growth image is as follows:

[0111] Step a: Model each stage of the crop pathological process through a convolutional neural network, obtain several pathological models, and generate a pathological model set = {pathological model 1, pathological model 2, ..., pathological model n};

[0112] Repeat step a, model the pathological process of different diseases of crops through convolutional neural network, obtain several pathological model sets, and generate plant abnormality set = {pathological model set 1, pathological model set 2, ..., pathological model set n};

[0113] Through the convolutional neural network, each stage of crop growth is modeled to obtain several stem and leaf models, and a plant morphology set is generated = {stem and leaf model 1, stem and leaf model 2, ..., stem and leaf model n};

[0114] Recognizing the growth image according to the plant morphology set, identifying the region of the crop in the growth image, and marking the region as a plant region to generate a plant region set;

[0115] A two-dimensional rectangular coordinate system A is established, and the growth image is mapped into the two-dimensional rectangular coordinate system A so that each pixel point in the growth image has its corresponding coordinate;

[0116] Get the coordinate with the smallest y value in the plant area and record it as the bottom line coordinate;

[0117] Get the minimum y value of all plant areas in the plant area set to generate a level set; get the median in the level set and mark it as the level value;

[0118] Set y = horizontal value (x∈R) as the horizontal line of the plant;

[0119] Get the distance between the bottom line coordinate and the horizontal line of the plant, and record it as the bottom line distance;

[0120] Set distance threshold;

[0121] If the bottom line distance > the distance threshold, the plant area is deleted from the plant area set;

[0122] If the bottom line distance is ≤ the distance threshold, no action is required;

[0123] identifying plant regions in the plant region set according to the plant anomaly set, and identifying diseased regions in the plant region indicating crop disease;

[0124] If there is a diseased area in the plant area, the plant area is marked as a diseased plant; if there is no diseased area in the plant area, no operation is required;

[0125] Specifically, the process of obtaining the plant distance according to the growth image and generating a temporary path node according to the plant distance includes:

[0126] Set the bottom threshold BT;

[0127] Establish a two-dimensional rectangular coordinate system B, and map the growth image into the two-dimensional rectangular coordinate system B so that each pixel in the growth image has its corresponding coordinate;

[0128] Get the minimum y value in the plant area and mark it as the bottom y value;

[0129] Obtain the top y value according to the bottom y value, wherein the top y value=bottom y value+bottom threshold BT;

[0130] Generate the bottom region in the two-dimensional rectangular coordinate system B according to the bottom y value and the top y value: bottom y value ≤ y ≤ top y value (x∈R);

[0131] Acquire the area where the plant area overlaps with the bottom area, and mark the area as the main stem area of ​​the plant;

[0132] Acquire stem images and leaf images at various stages of crop growth and generate several stem-leaf image pairs [stem image, leaf image];

[0133] It should be noted that the stem-leaf image pair is a stem image and a leaf image at the same stage, and the stem image only contains the stem of the crop, and the leaf image only contains the leaf of the crop;

[0134] Step b1: Obtain the pixel value corresponding to each pixel point in the stem image, generate a stem pixel set, compare the pixel values ​​in the stem pixel set, obtain the maximum value and the minimum value, mark the maximum value as the stem peak value, and mark the minimum value as the stem valley value; similarly, obtain the leaf pixel set, leaf peak value and leaf valley value of the leaf image;

[0135] Step b2: Generate a stem color value range [stem valley value, stem peak value] according to the stem peak value and the stem valley value, and generate a leaf color value range [leaf valley value, leaf peak value] according to the leaf peak value and the leaf valley value; if there is an overlapping area in the stem color value range and the leaf color value range in terms of value, then remove the overlapping area in the stem color value range and the leaf color value range,

[0136] Step b3: Obtain the mode of the pixel values ​​in the stem pixel set that belong to the stem color value range, recorded as the stem color mode; similarly, obtain the leaf color mode; obtain

[0137] Step b4: generating stem and leaf characteristic factors according to the stem color range, the leaf color range and the stem and leaf contrast value;

[0138] Repeat steps b1 to b4 to obtain the stem-leaf feature factors of the stem-leaf image pairs at each stage, and generate a stem-leaf feature set = {stem-leaf feature factor 1 (stem color range 1, leaf color range 1, stem-leaf contrast value 1), stem-leaf feature factor 2 (stem color range 2, leaf color range 2, stem-leaf contrast value 2), ..., stem-leaf feature factor n (Stem color range n , leaf color range n , stem-leaf ratio n )};

[0139] Set the light error threshold RET, and the light error threshold RET is a percentage;

[0140] Obtain the pixel value of each pixel point in the main stem area of ​​the plant to generate a pixel set in the bottom area;

[0141] Step c1: Filter the pixel values ​​in the bottom area pixel set, filter out the pixel values ​​whose pixel values ​​∈ the stem color value range, and generate the bottom stem pixel set according to the pixel values, and then filter out the pixel values ​​whose pixel values ​​∈ the leaf color value range, and generate the bottom leaf pixel set according to the pixel values;

[0142] Step c2: Obtain the majority of the bottom stem pixel set, recorded as the bottom stem majority; obtain the majority of the bottom leaf pixel set, recorded as the bottom leaf majority; generate the bottom stem majority and the bottom leaf majority according to the bottom stem majority and the bottom leaf majority. If [stem-leaf contrast value × (1-RET)] ≤ bottom stem-leaf contrast value ≤ [stem-leaf contrast value × (1+RET)], then the bottom stem mode is marked as the stem color value; otherwise, no operation is required;

[0143] According to the stem-leaf feature set, steps c1 and c2 are repeated. If a stem color value is obtained, it is marked as a characteristic color value; if several stem color values ​​are obtained, a stem color value is randomly selected and marked as a characteristic color value; if no stem color value is obtained, a warning message is generated to inform the user that the recognition fails;

[0144] Step d1: obtaining the coordinates corresponding to the characteristic color value of the pixel points in the main stem area of ​​the plant, marking them as stem coordinates, and generating a stem coordinate set;

[0145] Step d2: Obtain the coordinate with the largest x value in the stem coordinate set, marked as the stem X peak coordinate; obtain the coordinate with the smallest x value in the stem coordinate set, marked as the stem X valley coordinate; obtain the coordinate with the largest y value in the stem coordinate set, marked as the stem Y peak coordinate; obtain the coordinate with the smallest y value in the stem coordinate set, marked as the stem Y valley coordinate; generate the stem center coordinate according to the stem X peak coordinate, the stem X valley coordinate, the stem Y peak coordinate and the stem Y valley coordinate, and the stem center coordinate is expressed as follows:

[0146]

[0147] Repeat steps d1 and d2 to obtain the stem center coordinates of each plant region in the plant region set to generate a stem center coordinate set;

[0148] Obtain the adjacent stem center coordinates according to the x value, and generate a stem center coordinate pair (stem center coordinate A, stem center coordinate B);

[0149] Get stem distance according to stem center coordinate pair = |x 茎中心坐标A -x 茎中心坐标B |;

[0150] Set the road width threshold RWT;

[0151] If the stem distance < the road width threshold RWT, no action is required;

[0152] If the stem distance ≥ the road width threshold RWT, obtain the temporary path longitudinal line (y∈R); obtaining the growth image ratio and the image landmark corresponding to the growth image, and generating a temporary path node according to the temporary path longitudinal line, the growth image ratio and the image landmark, wherein the temporary path node is composed of longitude and latitude;

[0153] Specifically, the process of generating a harvesting path according to the departure place, destination, temporary path node and path data includes:

[0154] Establish a geographic coordinate system and map the path nodes, basic paths and temporary path nodes to the geographic coordinate system;

[0155] Get the growth image of the generated temporary path nodes and mark it as the path image;

[0156] Obtain an image number corresponding to the path image, obtain a processor that generates the image number, and query an instrument orientation of the processor, wherein the instrument orientation includes a left side of the irrigation agricultural machine and a right side of the irrigation agricultural machine;

[0157] If the instrument is oriented to the left of the irrigation machine, the temporary path node is taken as the origin, perpendicular to the basic path where the temporary path node is located, and extended to the left to the adjacent path node or basic path to generate a temporary path;

[0158] If the instrument is on the right side of the irrigation machine, the temporary path node is taken as the origin, perpendicular to the basic path where the temporary path node is located, and extends rightward to the adjacent path node or basic path to generate a temporary path.

[0159] Generate a shortest path from the departure point to the destination according to the departure point, the destination, the temporary path node, the temporary path, the path node and the basic path, and mark the shortest path as a harvesting path;

[0160] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An agricultural machinery management and control system based on Beidou navigation and positioning technology, including a control center, characterized in that: The control center is communicatively connected with a path data acquisition module, an irrigation path generation module, a growth image acquisition module and a harvesting path generation module; The path data acquisition module is used to set a Beidou short message device on the sowing agricultural machine to obtain the path data of the farmland through the sowing agricultural machine, and the path data includes path nodes and basic paths; The irrigation path generation module is used to obtain the irrigation radius of the irrigation agricultural machinery and generate the irrigation path according to the irrigation radius and path data; The growth image acquisition module is used to set the irrigation cycle. The irrigation agricultural machine irrigates the crops through the irrigation path according to the irrigation cycle and obtains the growth image of the crops; it is judged whether there are diseased plants in the crops according to the growth image. If there are diseased plants, the harvesting position is obtained; if there are no diseased plants, no operation is required; The harvesting path generation module is used to obtain the plant distance according to the growth image, and generate a temporary path node according to the plant distance; obtain the geographical location of the harvesting agricultural machinery through Beidou positioning, and record it as the departure place; obtain the harvesting position, and record it as the destination; generate a harvesting path according to the departure place, the destination, the temporary path node and the path data, and the Beidou navigation guides the harvesting agricultural machinery to harvest the diseased plants according to the harvesting path; The process of obtaining the irrigation radius of the irrigation agricultural machinery and generating the irrigation path according to the irrigation radius and path data includes: Acquire the irrigation radius of the irrigation agricultural machinery, wherein the irrigation radius includes the left irrigation radius and the right irrigation radius; simulate the basic driving path of the irrigation agricultural machinery, and acquire the unilateral irrigation range of the basic path according to the irrigation radius of the irrigation agricultural machinery, wherein the unilateral irrigation range includes the left unilateral irrigation range and the right unilateral irrigation range; The unilateral irrigation overlap rate of each basic path is obtained according to the unilateral irrigation range of each basic path. The unilateral irrigation overlap rate includes the left unilateral irrigation overlap rate and the right unilateral irrigation overlap rate. The unilateral irrigation overlap rate is the ratio of the overlapping range of the unilateral irrigation range of the basic path and the unilateral irrigation range of other basic paths to the unilateral irrigation range of the basic path, which is expressed as follows: Set the overlap threshold OLP; If the left unilateral irrigation overlap rate and the right unilateral irrigation overlap rate of basic path A are both greater than the overlap threshold OLP, no operation is required, that is, basic path A is not used as an irrigation path; otherwise, basic path A is marked as an irrigation path; The process of generating a harvesting path based on the departure point, destination, temporary path nodes and path data includes: Establish a geographic coordinate system and map the path nodes, basic paths and temporary path nodes to the geographic coordinate system; Get the growth image of the generated temporary path nodes and mark it as the path image; Obtain an image number corresponding to the path image, obtain a processor that generates the image number, and query an instrument orientation of the processor, wherein the instrument orientation includes a left side of the irrigation agricultural machine and a right side of the irrigation agricultural machine; If the instrument is on the left side of the irrigation machine, the temporary path node is taken as the origin, perpendicular to the basic path where the temporary path node is located, and extended to the adjacent path node or basic path to generate a temporary path; if the instrument is on the right side of the irrigation machine, the temporary path node is taken as the origin, perpendicular to the basic path where the temporary path node is located, and extended to the adjacent path node or basic path to generate a temporary path; The shortest path from the departure point to the destination is generated according to the departure point, the destination, the temporary path node, the temporary path, the path node and the basic path, and the shortest path is marked as a harvesting path.

2. The agricultural machinery management and control system based on Beidou navigation and positioning technology according to claim 1 is characterized in that: The process of setting up Beidou short message equipment on a sowing agricultural machine and obtaining the path data of the farmland through the sowing agricultural machine includes: A Beidou short message device is set on the seeding agricultural machinery to obtain the driving data of the seeding agricultural machinery in real time through Beidou positioning. The driving data includes the driving position and driving time. The driving position is composed of longitude and latitude. The driving position and driving time are in a one-to-one correspondence. The driving data is packaged and sent to the Beidou short message device, and the driving data is transmitted to the path data collection module through the Beidou short message device; The path data acquisition module generates a driving path according to the driving position and driving time in the driving data, locates the driving position in order from morning to night according to its corresponding driving time, and connects the locations in sequence to generate a driving path; The driving positions in the three driving data are obtained in chronological order, which are recorded as D1 (LON1, LAT1), D2 (LON2, LAT2) and D3 (LON3, LAT3), where LON is longitude and LAT is latitude; Set the error threshold LLE; Get standard longitude and standard latitude like Then the driving position D2 is marked as a path node; otherwise, no operation is required; If there is a driving path intersection in the driving path, the intersection position of the driving path is obtained, and the intersection position is marked as a path node; If there is a driving path between the path nodes, it is marked as a basic path.

3. The agricultural machinery management and control system based on Beidou navigation and positioning technology according to claim 2 is characterized in that: The process of obtaining the growth image of the crops and determining whether there are diseased plants in the crops according to the growth image, and if there are diseased plants, obtaining the harvesting position includes: Cameras and processors are arranged on the left and right sides of the irrigation machine; Set the image acquisition cycle IAC; The camera is used to obtain growth images of crops according to an image acquisition cycle IAC; The processor is used to generate a unique image number based on the growth image, and obtain the current driving position through Beidou positioning, which is recorded as an image landmark; the growth image, the image number and the image landmark are in a one-to-one correspondence with each other; at the same time, it is used to detect in real time whether there are diseased plants in the crops in the growth image. If there are diseased plants, the image landmark corresponding to the growth image is obtained, and the image landmark is marked as the harvesting position.

4. The agricultural machinery management and control system based on Beidou navigation and positioning technology according to claim 3 is characterized in that: The process of detecting whether there are diseased plants in the crops in the growth image includes: Step a: Model each stage of the crop pathological process through a convolutional neural network, obtain several pathological models, and generate a pathological model set; Repeat step a, model the pathological process of different diseases of crops through convolutional neural network, obtain several pathological model sets, and generate plant abnormality sets; Through convolutional neural networks, various stages of crop growth are modeled, several stem and leaf models are obtained, and a plant morphology set is generated; Recognizing the growth image according to the plant morphology set, identifying the region of the crop in the growth image, and marking the region as a plant region to generate a plant region set; Establish a two-dimensional rectangular coordinate system A, and map the growth image into the two-dimensional rectangular coordinate system A; Get the coordinate with the smallest y value in the plant area and record it as the bottom line coordinate; Get the minimum y value of all plant areas in the plant area set to generate a level set; get the median in the level set and mark it as the level value; In the two-dimensional rectangular coordinate system A, when x∈R, set the y=horizontal value to the horizontal line of the plant; Get the distance between the bottom line coordinate and the horizontal line of the plant, and record it as the bottom line distance; Set distance threshold; If the bottom line distance > the distance threshold, the plant area is deleted from the plant area set; If the bottom line distance is ≤ the distance threshold, no action is required; identifying plant regions in the plant region set according to the plant anomaly set, and identifying diseased regions in the plant region indicating crop disease; If there is a diseased area in the plant area, the plant area is marked as a diseased plant; if there is no diseased area in the plant area, no operation is required.

5. The agricultural machinery management and control system based on Beidou navigation and positioning technology according to claim 4 is characterized in that: The process of obtaining the plant distance according to the growth image and generating a temporary path node according to the plant distance includes: Set the bottom threshold BT; Establish a two-dimensional rectangular coordinate system B, and map the growth image into the two-dimensional rectangular coordinate system B; Get the minimum y value in the plant area and mark it as the bottom y value; Obtain the top y value according to the bottom y value, wherein the top y value=bottom y value+bottom threshold BT; Generate a bottom region in a two-dimensional rectangular coordinate system B according to the bottom y value and the top y value: in the two-dimensional rectangular coordinate system B, when x∈R, the bottom y value ≤ y ≤ the top y value; obtain an area where the plant area overlaps with the bottom region, and mark the area as the main stem area of ​​the plant; Acquire stem images and leaf images at various stages of crop growth to generate a number of stem-leaf image pairs [stem image, leaf image]; the stem-leaf image pair is a stem image and a leaf image at the same stage, and the stem image only contains the stem of the crop, and the leaf image only contains the leaf of the crop; Step b1: Obtain the pixel value corresponding to each pixel point in the stem image, generate a stem pixel set, compare the pixel values ​​in the stem pixel set, obtain the maximum value and the minimum value, mark the maximum value as the stem peak value, and mark the minimum value as the stem valley value; similarly, obtain the leaf pixel set, leaf peak value and leaf valley value of the leaf image; Step b2: Generate a stem color value range [stem valley value, stem peak value] according to the stem peak value and the stem valley value, and generate a leaf color value range [leaf valley value, leaf peak value] according to the leaf peak value and the leaf valley value; if the stem color value range and the leaf color value range have overlapping areas in terms of values, remove the overlapping areas in the stem color value range and the leaf color value range, so that Step b3: Obtain the mode of the pixel values ​​in the stem pixel set that belong to the stem color value range, recorded as the stem color mode; similarly, obtain the leaf color mode; according to Step b4: generating stem and leaf characteristic factors according to the stem color range, the leaf color range and the stem and leaf contrast value; Repeat steps b1 to b4 to obtain stem-leaf feature factors of the stem-leaf image pairs at each stage and generate a stem-leaf feature set; Set the light error threshold RET, and the light error threshold RET is a percentage; Obtain the pixel value of each pixel point in the main stem area of ​​the plant to generate a pixel set in the bottom area; Step c1: Filter the pixel values ​​in the bottom area pixel set, filter out the pixel values ​​whose pixel values ​​∈ the stem color value range, and generate the bottom stem pixel set according to the pixel values, and then filter out the pixel values ​​whose pixel values ​​∈ the leaf color value range, and generate the bottom leaf pixel set according to the pixel values; Step c2: Get the mode of the bottom stem pixel set, recorded as the bottom stem mode; get the mode of the bottom leaf pixel set, recorded as the bottom leaf mode; according to If [stem-leaf contrast value × (1-RET)] ≤ bottom stem-leaf contrast value ≤ [stem-leaf contrast value × (1+RET)], then the bottom stem mode is marked as the stem color value; otherwise, no operation is required; According to the stem-leaf feature set, steps c1 and c2 are repeated. If a stem color value is obtained, it is marked as a characteristic color value; if several stem color values ​​are obtained, a stem color value is randomly selected and marked as a characteristic color value; if no stem color value is obtained, a warning message is generated to inform the user that the recognition fails; Step d1: obtaining the coordinates corresponding to the characteristic color value of the pixel points in the main stem area of ​​the plant, marking them as stem coordinates, and generating a stem coordinate set; Step d2: Obtain the coordinate with the largest x value in the stem coordinate set, marked as the stem X peak coordinate; obtain the coordinate with the smallest x value in the stem coordinate set, marked as the stem X valley coordinate; obtain the coordinate with the largest y value in the stem coordinate set, marked as the stem Y peak coordinate; obtain the coordinate with the smallest y value in the stem coordinate set, marked as the stem Y valley coordinate; generate the stem center coordinate according to the stem X peak coordinate, the stem X valley coordinate, the stem Y peak coordinate and the stem Y valley coordinate, and the stem center coordinate is expressed as follows: Repeat steps d1 and d2 to obtain the stem center coordinates of each plant region in the plant region set to generate a stem center coordinate set; Obtain the adjacent stem center coordinates according to the x value, and generate a stem center coordinate pair (stem center coordinate A, stem center coordinate B); Get stem distance according to stem center coordinate pair = |x 茎中心坐标A -x 茎中心坐标B |; Set the road width threshold RWT; If the stem distance < the road width threshold RWT, no action is required; If the stem distance ≥ the road width threshold RWT, then in the two-dimensional rectangular coordinate system B, in the case of y∈R, obtain the temporary path vertical line The growth image ratio and the image landmark corresponding to the growth image are obtained, and a temporary path node is generated according to the temporary path longitudinal line, the growth image ratio and the image landmark, wherein the temporary path node is composed of longitude and latitude.

Images