CN117197655A - Rice leaf roller hazard degree prediction method, device, electronic equipment and medium - Google Patents
Rice leaf roller hazard degree prediction method, device, electronic equipment and medium Download PDFInfo
- Publication number
- CN117197655A CN117197655A CN202310961079.3A CN202310961079A CN117197655A CN 117197655 A CN117197655 A CN 117197655A CN 202310961079 A CN202310961079 A CN 202310961079A CN 117197655 A CN117197655 A CN 117197655A
- Authority
- CN
- China
- Prior art keywords
- disease
- blades
- rice
- total number
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 81
- 241000008892 Cnaphalocrocis patnalis Species 0.000 title claims abstract description 58
- 238000005070 sampling Methods 0.000 claims abstract description 254
- 201000010099 disease Diseases 0.000 claims abstract description 193
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 193
- 241000209094 Oryza Species 0.000 claims abstract description 111
- 235000007164 Oryza sativa Nutrition 0.000 claims abstract description 111
- 235000009566 rice Nutrition 0.000 claims abstract description 111
- 230000006378 damage Effects 0.000 claims abstract description 30
- 230000011218 segmentation Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000005096 rolling process Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 4
- 241000255901 Tortricidae Species 0.000 abstract description 21
- 238000004220 aggregation Methods 0.000 abstract description 8
- 230000002776 aggregation Effects 0.000 abstract description 8
- 238000009826 distribution Methods 0.000 description 7
- 241000607479 Yersinia pestis Species 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 241000238631 Hexapoda Species 0.000 description 4
- 208000015181 infectious disease Diseases 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000032669 eclosion Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000575 pesticide Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method, a device, electronic equipment and a medium for predicting the damage degree of rice leaf rollers, and relates to the field of farmland disease prediction, wherein the method comprises the following steps: determining the total number of the first disease blades according to the total number of each global blade corresponding to all the non-random local sampling areas in the area to be predicted and a preset quadratic linear equation; obtaining the number of second disease blades corresponding to the random local sampling area according to the number of the global total blades corresponding to the random local sampling area and a preset one-time equation so as to obtain the total number of second disease blades corresponding to all the random local sampling areas; and determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the global total blades of each prediction subarea. The method fully considers the local aggregation and the spatial randomness of the rice leaf rollers, and can accurately estimate the leaf roller rate caused by the leaf rollers, thereby predicting the damage degree of the rice leaf rollers.
Description
Technical Field
The invention relates to the field of farmland disease prediction, in particular to a method, a device, electronic equipment and a medium for predicting the damage degree of rice leaf rollers.
Background
The method for predicting the damage degree of the leaf rollers usually adopts an inverse distance weight method, and the distribution of the leaf roller diseases and insect pests often has local aggregation and spatial randomness, so that the accuracy of the estimated damage degree is greatly reduced.
Disclosure of Invention
The invention provides a rice leaf roller hazard degree prediction method, a device, electronic equipment and a medium, which are used for solving the technical problem that the existing rice leaf roller hazard degree prediction method is inaccurate.
In a first aspect, the invention provides a method for predicting the hazard degree of rice leaf rollers, which comprises the following steps:
determining the total number of first disease blades corresponding to all non-random local sampling areas according to the total number of each global blade corresponding to all non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, wherein the area to be predicted comprises all non-random local sampling areas and all random local sampling areas;
for each random local sampling area, acquiring the number of second disease blades corresponding to the random local sampling area according to the number of global total blades corresponding to the random local sampling area and a preset one-time equation so as to acquire the total number of second disease blades corresponding to all the random local sampling areas;
Determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the blades of each prediction subarea;
the preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area;
the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice;
the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
According to the rice leaf roller hazard degree prediction method provided by the invention, before determining the total number of first disease leaves corresponding to all the non-random local sampling areas, the method further comprises the following steps:
Above the rice in the area to be predicted, sampling by using the unmanned aerial vehicle at a second preset angle at a center position spaced by a second preset distance from the rice to obtain images of all sampling areas;
splicing all the sampling area images to form an initial field area image;
and carrying out coordinate correction on the initial field area image according to the longitude and latitude of the field edge corner of each sampling area image, and obtaining the area image to be predicted.
According to the rice leaf roller hazard degree prediction method provided by the invention, after the image of the area to be predicted is acquired, the method further comprises the following steps:
dividing the region image to be predicted to obtain all sub-images to be predicted;
inputting all the sub-images to be predicted into a preset blade prediction model, and obtaining the total number of predicted blades corresponding to each sub-image to be predicted, which is output by the preset blade prediction model;
the preset blade prediction model is determined according to sample prediction images and sample blade total number training.
According to the rice leaf roller hazard degree prediction method provided by the invention, after the total number of predicted blades corresponding to each sub-image to be predicted output by the preset blade prediction model is obtained, the method further comprises the following steps:
Determining the region where the sub-image to be predicted exists in the target position as a random local sampling region, and determining all random local sampling regions;
and determining the region where the sub-image to be predicted which does not exist in the target position is a non-random local sampling region, and determining all the non-random local sampling regions.
According to the rice leaf roller hazard degree prediction method provided by the invention, before determining the total number of first disease leaves corresponding to all the non-random local sampling areas, the method further comprises the following steps:
sampling at a first preset angle by using an unmanned plane at any target position above the rice at intervals of a first preset distance of the rice to obtain all local sampling images;
determining all overlapped area images according to all local sampling images and all random local sampling areas;
inputting all the images of the overlapped areas to a semantic segmentation model, and obtaining the output of the semantic segmentation model, wherein the number of the local disease blades and the number of the local total blades corresponding to each overlapped area;
the semantic segmentation model is determined according to sample disease images, the number of sample disease blades and the training of the number of sample local total blades.
According to the method for predicting the damage degree of the rice leaf rollers, which is provided by the invention, the total number of first disease blades corresponding to all the non-random local sampling areas is determined according to the total number of all the global blades corresponding to all the non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, and the method comprises the following steps:
inputting the total number of each global blade to the preset quadratic linear equation, and obtaining the total number of disease blades corresponding to each non-random local sampling area;
and determining the total number of the first disease blades according to the sum of the total number of the disease blades corresponding to all the non-random local sampling areas.
According to the method for predicting the damage degree of the rice leaf rollers, the second disease blade number corresponding to the random local sampling area is obtained according to the total global blade number corresponding to the random local sampling area and a preset one-time equation, so as to obtain the total second disease blade number corresponding to all the random local sampling areas, and the method comprises the following steps:
inputting the total number of the global blades corresponding to each random local sampling area to a preset one-time equation corresponding to the random local sampling area, and obtaining the total number of the disease blades corresponding to each random local sampling area, which is output by the preset one-time equation;
And determining the total number of the second disease blades according to the sum of the total number of the disease blades corresponding to all the random local sampling areas.
According to the method for predicting the damage degree of the rice leaf rollers, which is provided by the invention, the prediction result of the rice diseases is determined according to the total number of the first disease blades, the total number of the second disease blades and the total number of the global total blades of each prediction subarea, and the method comprises the following steps:
determining the total number of the rice disease leaves in the area to be predicted according to the sum of the total number of the first disease leaves and the total number of the second disease leaves;
determining the total number of blades of the area to be predicted according to the sum of the total number of blades of the overall situation of each prediction subarea;
determining the leaf rolling rate of the rice diseases according to the total number of the rice disease leaves in the area to be predicted and the quotient of the total number of the leaves in the area to be predicted;
and determining the rice disease prediction result of the region to be predicted from a preset rice hazard degree table according to the rice disease leaf rolling rate.
In a second aspect, a device for predicting damage degree of rice leaf rollers is provided, including:
the first determining unit is used for determining the total number of first disease blades corresponding to all the non-random local sampling areas according to the total global total blade number corresponding to all the non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, and the area to be predicted comprises all the non-random local sampling areas and all the random local sampling areas;
The acquisition unit is used for acquiring the number of second disease blades corresponding to the random local sampling areas according to the number of global total blades corresponding to the random local sampling areas and a preset equation for each random local sampling area so as to acquire the total number of second disease blades corresponding to all the random local sampling areas;
the second determining unit is used for determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the global total blades of each prediction subarea;
the preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area;
the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice;
the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements any one of the above methods for predicting the hazard level of rice leaf rollers when executing the program.
In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a rice leaf roller hazard level prediction method as described in any one of the above.
The invention provides a rice leaf roller hazard degree prediction method, a device, electronic equipment and a medium, wherein a related preset equation is constructed according to disease leaves and total leaves acquired from local collection through the combination of the local collection and global collection, and the disease leaves are calculated according to different preset equations by combining the total leaves acquired from the global collection according to different collection areas, and a rice disease prediction result is determined according to the total number of the disease leaves and the total leaves.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for predicting the hazard degree of rice leaf rollers;
FIG. 2 is a second flow chart of the method for predicting the damage degree of rice leaf rollers provided by the invention;
FIG. 3 is a third flow chart of the method for predicting the damage degree of rice leaf rollers provided by the invention;
FIG. 4 is a flow chart of a method for predicting the hazard degree of rice leaf rollers, which is provided by the invention;
FIG. 5 is a fifth flow chart of the method for predicting the hazard degree of rice leaf rollers provided by the invention;
FIG. 6 is a schematic diagram of dividing a region to be predicted according to the present invention;
FIG. 7 is a schematic structural view of the rice leaf roller hazard degree prediction device provided by the invention;
fig. 8 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The rice leaf rollers are serious insect pests in rice production and are widely distributed in various rice planting areas, the rice leaf rollers form continuous harm to rice leaves through multiple generations and high-frequency spawning and eclosion, the rice yield is finally influenced, and the main means for the hazard control of leaf rollers at the present stage is to spray proper amounts of pesticides according to different hazard degrees, so that the hazard degree prediction of leaf rollers is necessary.
For predicting the leaf roller hazard degree, the current method mainly comprises the steps of manually collecting data and estimating by using an interpolation algorithm, observing the hazard degree of part of the leaf rollers of the paddy field, and predicting the hazard degree of the leaf rollers of the paddy field in large blocks by using a spatial interpolation algorithm, for example, a reverse distance weighting method. The partial area in the paddy field is selected randomly as initial data, the initial data are processed, and the damage degree of the plant diseases and insect pests of the area is estimated through an interpolation algorithm. The interpolation algorithm mainly used is an inverse distance weighting method, a kriging method, a spline function method, a polynomial interpolation method and the like.
The method is generally used for predicting the damage degree of the leaf rollers, and the distribution of the leaf rollers and insect pests often has local aggregation and spatial randomness, so that the accuracy of the estimated damage degree is greatly reduced; another approach is based on manually collecting the process data and using spline algorithms that are difficult to estimate the error, resulting in a significant drop in accuracy, while also failing to adequately account for the local aggregations and spatial randomness of the diseases of volume She Mingchong. In order to overcome the defects of the prior art, the invention provides a rice leaf roller hazard degree prediction method, a device, electronic equipment and a medium, and fig. 1 is one of flow diagrams of the rice leaf roller hazard degree prediction method provided by the invention, wherein the rice leaf roller hazard degree prediction method comprises the following steps:
and step 101, determining the total number of first disease blades corresponding to all the non-random local sampling areas according to the total number of all the global total blades corresponding to all the non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, wherein the area to be predicted comprises all the non-random local sampling areas and all the random local sampling areas.
In step 101, the area to be predicted includes all non-random local sampling areas and all random local sampling areas, by spacing any target position of a first preset distance of rice above the rice, sampling at a first preset angle, for example, randomly selecting P target paddy field points in random sampling, using an unmanned plane to photograph the condition of the rice obliquely downwards at an angle of 45 ° at a position 10 cm above the tip of the rice, acquiring a sampling image and using a satellite to position and record the position, accumulating the acquisition times for P times, thereby determining the adopted area corresponding to the target position as a random local sampling area, and removing other areas except all random local sampling areas as all non-random local sampling areas, wherein the set of all non-random local sampling areas and all random local sampling areas is the area to be predicted.
Optionally, the number of each global total blade corresponding to all the non-random local sampling areas is determined according to a central position above the rice in the area to be predicted, which is spaced by a second preset distance from the rice, and sampling is performed at a second preset angle, the first preset distance is smaller than the second preset distance, the first preset angle is smaller than the second preset angle, for example, in global sampling, an angle of 500 cm above the blade tip and 90 ° vertically downward is used by an unmanned aerial vehicle, sampling is performed on a target paddy field at a certain resolution, and a sampling image is obtained and a satellite positioning position is recorded.
Optionally, the preset quadratic equation is generated by fitting the number of local disease blades and the number of local total blades in all random local sampling areas, and the number of local disease blades and the number of local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice, so that the invention analyzes the rice image obtained by random sampling to determine the number of local disease blades and the number of local total blades, fits the number of local disease blades and the number of local total blades in all random local sampling areas, determines the relevant parameters of the quadratic equation, and constructs the preset quadratic equation, for example, the preset quadratic equation is:
y＝ax 2 +bx+c (1)
in the formula (1), the parameters a, b and c are determined by fitting according to the number of the local disease blades and the number of the local total blades in all the random local sampling areas.
Optionally, the determining the total number of the first disease blades corresponding to all the non-random local sampling areas according to the total global blade number corresponding to all the non-random local sampling areas in the area to be predicted and the preset quadratic linear equation includes:
Inputting the total number of each global blade to the preset quadratic linear equation, and obtaining the total number of disease blades corresponding to each non-random local sampling area;
and determining the total number of the first disease blades according to the sum of the total number of the disease blades corresponding to all the non-random local sampling areas.
Optionally, substituting each global total blade number as x into formula (1), and obtaining y, where y is the total number of disease blades corresponding to each non-random local sampling area, and determining the total number of first disease blades according to the sum of the total number of disease blades corresponding to all non-random local sampling areas, where the following formula may be referred to:
in the formula (2), harm u D is the number of all non-random local sampling areas and z is the total number of the first disease blades i For each global total number of blades.
Step 102, for each random local sampling area, according to the total global blade number corresponding to the random local sampling area and a preset equation, obtaining the second disease blade number corresponding to the random local sampling area, so as to obtain the total second disease blade number corresponding to all the random local sampling areas.
In step 102, each of the predetermined equations is determined according to the number of local disease blades and the number of local total blades in each of the random local sampling regions, and the present invention determines the coefficient q of each random local sampling region according to the quotient of the number of local disease blades and the number of local total blades in each of the random local sampling regions k Thus, for each random local sampling region, a preset one-time equation is constructed, and the formula can be referred to:
m＝q k n (3)
in formula (3), q k And for the coefficient of each random local sampling area, the value of k is 1 to e, and e is the number of areas of the random local sampling area.
Optionally, the number of global total blades corresponding to the random local sampling area is determined by sampling at a second preset angle from a center position above the rice in the area to be predicted at a second preset distance from the rice, and after determining the coefficient of each random local sampling area and the number of global total blades corresponding to each random local sampling area, the invention determines the number of second disease blades corresponding to each random local sampling area by adopting different coefficients of random local sampling areas according to different numbers of global total blades corresponding to different random local sampling areas.
Optionally, the obtaining the number of the second disease blades corresponding to the random local sampling area according to the number of the global total blades corresponding to the random local sampling area and a preset one-time equation to obtain the total number of the second disease blades corresponding to all random local sampling areas includes:
Inputting the total number of the global blades corresponding to each random local sampling area to a preset one-time equation corresponding to the random local sampling area, and obtaining the total number of the disease blades corresponding to each random local sampling area, which is output by the preset one-time equation;
and determining the total number of the second disease blades according to the sum of the total number of the disease blades corresponding to all the random local sampling areas.
Alternatively, the present invention may refer to the following formula:
in formula (4), harm v For the total number of the second disease leaves, n k Global total blade number, q, corresponding to each random local sampling area k And for the coefficient of each random local sampling area, the value of k is 1 to e, and e is the number of areas of the random local sampling area.
And step 103, determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the global total blades of each prediction subarea.
In step 103, the determining a prediction result of the rice disease according to the total number of the first disease blades, the total number of the second disease blades and the total number of the global total blades of each prediction subarea includes:
determining the total number of the rice disease leaves in the area to be predicted according to the sum of the total number of the first disease leaves and the total number of the second disease leaves;
Determining the total number of blades of the area to be predicted according to the sum of the total number of blades of the overall situation of each prediction subarea;
determining the leaf rolling rate of the rice diseases according to the total number of the rice disease leaves in the area to be predicted and the quotient of the total number of the leaves in the area to be predicted;
and determining the rice disease prediction result of the region to be predicted from a preset rice hazard degree table according to the rice disease leaf rolling rate.
Optionally, the method includes that the total number of the first disease leaves and the total number of the second disease leaves are added to determine the total number of the rice leaves in the area to be predicted, then the total number of the total leaves in each prediction subarea is summed to determine the total number of the leaves in the area to be predicted, then the total number of the rice leaves in the area to be predicted is divided by the total number of the leaves in the area to be predicted to determine the leaf rolling rate of the rice leaves, and finally according to the leaf rolling rate of the rice leaves, a rice disease prediction result of the area to be predicted is determined from a preset rice hazard degree table, optionally, for example, the preset rice hazard degree table comprises non-hazard risks corresponding to a first value range, mild hazard risks corresponding to a second value range, moderate hazard risks corresponding to a third value range, and serious hazard risks corresponding to a fourth value range, wherein the first value range is smaller than the second value range, the second value range is smaller than the third value range, the third value range is smaller than the fourth value range, and the rice hazard prediction result of the rice leaves in the area to be predicted is determined.
Fig. 6 is a schematic diagram of the division of the region to be predicted, in which the distribution of leaf rollers has spatial variation, and in general, infection points are distributed in an aggregated manner, and random distribution is shown among different infection blocks. For this purpose, a categorization discussion is to be made. As shown in fig. 6, a plurality of points are randomly selected, and the paddy field is divided into a plurality of blocks, and then the paddy block is divided into a random local sampling area and a non-random local sampling area according to whether the random sampling points exist in the block.
Alternatively, different types of rice blocks may require different fitting equations to be established for random local sampling regions as well as for non-random local sampling regions. For a random local sampling area, namely a rice block with analysis points, the rice leaf roller infection points are required to be considered to be distributed in an aggregation mode, so that a local fitting equation is established, and the point positions in the field are used for resolving, so that the estimation precision of the field is effectively improved; for a non-random local sampling area, namely a rice block without analysis points, the random distribution among different infected blocks of the rice leaf rollers is considered, so that a global fitting equation is established, and the global random points are used for resolving, so that the local aggregation and the spatial randomness of the rice leaf rollers are fully considered, the leaf rolling rate caused by the leaf rollers can be accurately estimated, and the damage degree of the rice leaf rollers is predicted.
The invention provides a rice leaf roller hazard degree prediction method, a device, electronic equipment and a medium, wherein a related preset equation is constructed according to disease leaves and total leaves acquired from local collection through the combination of the local collection and global collection, and the disease leaves are calculated according to different preset equations by combining the total leaves acquired from the global collection according to different collection areas, and a rice disease prediction result is determined according to the total number of the disease leaves and the total leaves.
Fig. 2 is a second flow chart of the method for predicting the damage degree of rice leaf rollers, which is provided by the invention, before determining the total number of first disease leaves corresponding to all non-random local sampling areas, the method further comprises:
step 201, sampling at a second preset angle by using an unmanned plane at a center position above the rice in the area to be predicted, wherein the center position is spaced by a second preset distance from the rice, and obtaining images of all sampling areas.
In step 201, the present invention may sample the target paddy field with a certain resolution at a position 500 cm above the blade tip and an angle of 90 ° vertically downward based on the data of the pest paddy field of the leaf roller collected by the unmanned aerial vehicle, obtain each sampling image, and record the satellite positioning position until traversing the region to be predicted, and obtain all the sampling region images.
And 202, splicing all the sampling area images to form an initial field area image.
In step 202, the present invention concatenates all of the sample area images based on the satellite positioning position of each of the sample images to form an initial field area image.
And 203, carrying out coordinate correction on the initial field area image according to the longitude and latitude of the field edge corner of each sampling area image, and obtaining the area image to be predicted.
In step 203, the present invention measures the coordinates of each sampling area image according to Differential GPS (Differential GPS-DGPS), forms detailed longitude and latitude including the corner of the field edge, uses data processing software Pix4D to splice each sampling area image, and uses the longitude and latitude of the field edge of each sampling area image to perform coordinate correction.
Fig. 3 is a third flow chart of the method for predicting the damage degree of rice leaf rollers, which is provided by the invention, after the image of the area to be predicted is obtained, the method further comprises:
and step 301, dividing the region image to be predicted, and obtaining all sub images to be predicted.
In step 301, the present invention divides the region image to be predicted obtained after all the sampling region images are spliced, and optionally, each sub-image to be predicted has the same size, i.e., the area of each sub-region after dividing the region to be predicted is the same.
Step 302, inputting all sub-images to be predicted into a preset blade prediction model, and obtaining the total number of predicted blades corresponding to each sub-image to be predicted, which is output by the preset blade prediction model;
the preset blade prediction model is determined according to sample prediction images and sample blade total number training.
In step 302, for each sub-image to be predicted, a deep learning algorithm is adopted to identify the total number of blades of each sub-image to be predicted, specifically, all sub-images to be predicted are input into a preset blade prediction model, and the total number of predicted blades corresponding to each sub-image to be predicted, which is output by the preset blade prediction model, is obtained.
The method is based on the rice field data of the leaf roller pests collected by the unmanned aerial vehicle, the deep learning is used for extracting the characteristics, the grid division is firstly carried out on the field of the image of the area to be predicted, and then the global and local fitting equations are used, so that the problem of low estimation space precision in the original method is solved.
Fig. 4 is a schematic flow chart of a method for predicting the damage degree of rice leaf rollers, which is provided by the invention, after obtaining the total number of predicted blades corresponding to each sub-image to be predicted output by the preset blade prediction model, the method further comprises:
And 401, determining the region where the sub-image to be predicted with the target position is located as a random local sampling region, and determining all random local sampling regions.
In step 401, after the total number of predicted blades corresponding to each sub-image to be predicted output by the preset blade prediction model is obtained, the classification of each sub-image to be predicted needs to be determined, and the invention determines whether the sub-image to be predicted is a random local sampling area or a non-random local sampling area according to whether a target position exists in the sub-image to be predicted, wherein the target position is a position above rice spaced by a first preset distance and sampled at a first preset angle, namely, the random local sampling area adopted by the invention is determined according to random sampling points, as shown in fig. 6.
And step 402, determining the area where the sub-image to be predicted which does not exist in the target position is a non-random local sampling area, and determining all the non-random local sampling areas.
In step 402, as shown in fig. 6, the sub-image to be predicted without random sampling points is a non-random local sampling region, and the present invention determines other sub-images to be predicted except all random local sampling regions as all non-random local sampling regions.
Fig. 5 is a fifth flow chart of a method for predicting the damage degree of rice leaf rollers, which is provided by the invention, before determining the total number of first disease leaves corresponding to all non-random local sampling areas, the method further comprises:
step 501, sampling at a first preset angle by using an unmanned plane at any target position above the rice at a first preset distance from the rice, and obtaining all local sampling images.
In step 501, the method adopts a random local sampling mode, samples at any target position above the rice at intervals of a first preset distance of the rice by using an unmanned plane at a first preset angle to acquire all local sampling images, optionally, the first preset distance is 10 cm, and the first preset angle is 45 degrees.
Step 502, determining all overlapped area images according to all local sampling images and all random local sampling areas.
In step 502, in an alternative embodiment, the local sampling image is located in any random local sampling area, where the overlapping area image is the local sampling image; if the local sampling image spans two adjacent random local sampling areas, the overlapping area images are two, the first is the overlapping part of the local sampling image and one random local sampling area, and the second is the overlapping part of the local sampling image and the other random local sampling area.
Step 503, inputting all the images of the overlapped areas into a semantic segmentation model, and obtaining the output of the semantic segmentation model, wherein the number of the local disease blades and the number of the local total blades corresponding to each overlapped area;
the semantic segmentation model is determined according to sample disease images, the number of sample disease blades and the training of the number of sample local total blades.
In step 503, the semantic segmentation model is determined according to the sample disease image, the number of sample disease blades and the number of sample local total blades, so that images of all overlapping areas are input into the semantic segmentation model, the semantic segmentation model is obtained and output, the number of local disease blades and the number of local total blades corresponding to each overlapping area are obtained.
Optionally, the leaf blades damaged by the leaf rollers are distributed in an aggregation manner in the paddy field, and if the leaf blades damaged by the leaf rollers exist somewhere, the probability of the leaf blades damaged by the leaf rollers existing nearby is higher, and from the perspective of the whole paddy field, the leaf blade distribution damaged by the leaf rollers has randomness. Therefore, the rice field points are randomly selected, the rice field blocks are divided, and the local solution and the global solution are used for classification calculation, so that the hazard degree of the leaf rollers is finally obtained, and the overlarge errors caused by the local aggregation and the spatial randomness of the leaf roller disease distribution can be avoided.
Fig. 7 is a schematic structural diagram of the rice leaf roller hazard degree prediction device provided by the invention, which comprises a first determining unit 1, wherein the first determining unit is used for determining the total number of first disease leaves corresponding to all non-random local sampling areas according to the total number of each global leaf corresponding to all non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, the area to be predicted comprises all non-random local sampling areas and all random local sampling areas, and the working principle of the first determining unit 1 can refer to the step 101 and is not repeated herein.
The rice leaf roller hazard degree prediction device further comprises an acquisition unit 2, wherein the acquisition unit is used for acquiring the number of second disease blades corresponding to the random local sampling areas according to the number of global total blades corresponding to the random local sampling areas and a preset equation for each random local sampling area so as to acquire the total number of second disease blades corresponding to all the random local sampling areas, and the working principle of the acquisition unit 2 can refer to the step 102 and is not repeated herein.
The rice leaf roller hazard degree prediction device further comprises a second determination unit 3, the second determination unit is configured to determine a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the global total blades in each prediction sub-region, and the working principle of the second determination unit 3 may refer to the foregoing step 103 and will not be repeated herein.
The preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area;
the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice;
the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
The invention provides a rice leaf roller hazard degree prediction method, a device, electronic equipment and a medium, wherein a related preset equation is constructed according to disease leaves and total leaves acquired from local collection through the combination of the local collection and global collection, and the disease leaves are calculated according to different preset equations by combining the total leaves acquired from the global collection according to different collection areas, and a rice disease prediction result is determined according to the total number of the disease leaves and the total leaves.
Fig. 8 is a schematic structural diagram of an electronic device provided by the present invention. As shown in fig. 8, the electronic device may include: processor 810, communication interface (Communications Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. Processor 810 can invoke logic instructions in memory 830 to perform a rice leaf roller hazard level prediction method comprising: determining the total number of first disease blades corresponding to all non-random local sampling areas according to the total number of each global blade corresponding to all non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, wherein the area to be predicted comprises all non-random local sampling areas and all random local sampling areas; for each random local sampling area, acquiring the number of second disease blades corresponding to the random local sampling area according to the number of global total blades corresponding to the random local sampling area and a preset one-time equation so as to acquire the total number of second disease blades corresponding to all the random local sampling areas; determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the blades of each prediction subarea; the preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area; the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice; the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, where the computer program product includes a computer program, where the computer program may be stored on a non-transitory computer readable storage medium, and when the computer program is executed by a processor, the computer is capable of executing the method, the apparatus, the electronic device, and the medium for predicting the hazard degree of rice leaf rollers provided by the above methods, where the method includes: determining the total number of first disease blades corresponding to all non-random local sampling areas according to the total number of each global blade corresponding to all non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, wherein the area to be predicted comprises all non-random local sampling areas and all random local sampling areas; for each random local sampling area, acquiring the number of second disease blades corresponding to the random local sampling area according to the number of global total blades corresponding to the random local sampling area and a preset one-time equation so as to acquire the total number of second disease blades corresponding to all the random local sampling areas; determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the blades of each prediction subarea; the preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area; the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice; the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
In still another aspect, the present invention provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the rice leaf roller hazard level prediction method provided by the above methods, the method comprising: determining the total number of first disease blades corresponding to all non-random local sampling areas according to the total number of each global blade corresponding to all non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, wherein the area to be predicted comprises all non-random local sampling areas and all random local sampling areas; for each random local sampling area, acquiring the number of second disease blades corresponding to the random local sampling area according to the number of global total blades corresponding to the random local sampling area and a preset one-time equation so as to acquire the total number of second disease blades corresponding to all the random local sampling areas; determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the blades of each prediction subarea; the preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area; the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice; the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (11)
1. The rice leaf roller hazard degree prediction method is characterized by comprising the following steps:
determining the total number of first disease blades corresponding to all non-random local sampling areas according to the total number of each global blade corresponding to all non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, wherein the area to be predicted comprises all non-random local sampling areas and all random local sampling areas;
for each random local sampling area, acquiring the number of second disease blades corresponding to the random local sampling area according to the number of global total blades corresponding to the random local sampling area and a preset one-time equation so as to acquire the total number of second disease blades corresponding to all the random local sampling areas;
Determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the blades of each prediction subarea;
the preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area;
the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice;
the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
2. The method for predicting the damage degree of rice leaf rollers according to claim 1, wherein before determining the total number of first disease leaves corresponding to all non-random local sampling areas, the method further comprises:
Above the rice in the area to be predicted, sampling by using the unmanned aerial vehicle at a second preset angle at a center position spaced by a second preset distance from the rice to obtain images of all sampling areas;
splicing all the sampling area images to form an initial field area image;
and carrying out coordinate correction on the initial field area image according to the longitude and latitude of the field edge corner of each sampling area image, and obtaining the area image to be predicted.
3. The method for predicting the damage degree of rice leaf rollers according to claim 2, wherein after acquiring the image of the area to be predicted, the method further comprises:
dividing the region image to be predicted to obtain all sub-images to be predicted;
inputting all the sub-images to be predicted into a preset blade prediction model, and obtaining the total number of predicted blades corresponding to each sub-image to be predicted, which is output by the preset blade prediction model;
the preset blade prediction model is determined according to sample prediction images and sample blade total number training.
4. The method for predicting the damage degree of rice leaf rollers according to claim 3, wherein after obtaining the total number of predicted leaves corresponding to each sub-image to be predicted output by the preset leaf prediction model, the method further comprises:
Determining the region where the sub-image to be predicted exists in the target position as a random local sampling region, and determining all random local sampling regions;
and determining the region where the sub-image to be predicted which does not exist in the target position is a non-random local sampling region, and determining all the non-random local sampling regions.
5. The method for predicting the damage degree of rice leaf rollers according to claim 1, wherein before determining the total number of first disease leaves corresponding to all non-random local sampling areas, the method further comprises:
sampling at a first preset angle by using an unmanned plane at any target position above the rice at intervals of a first preset distance of the rice to obtain all local sampling images;
determining all overlapped area images according to all local sampling images and all random local sampling areas;
inputting all the images of the overlapped areas to a semantic segmentation model, and obtaining the output of the semantic segmentation model, wherein the number of the local disease blades and the number of the local total blades corresponding to each overlapped area;
the semantic segmentation model is determined according to sample disease images, the number of sample disease blades and the training of the number of sample local total blades.
6. The method for predicting the damage degree of rice leaf rollers according to claim 1, wherein determining the total number of first disease leaves corresponding to all non-random local sampling areas according to the total number of each global leaf corresponding to all non-random local sampling areas in the area to be predicted and a preset quadratic linear equation comprises:
inputting the total number of each global blade to the preset quadratic linear equation, and obtaining the total number of disease blades corresponding to each non-random local sampling area;
and determining the total number of the first disease blades according to the sum of the total number of the disease blades corresponding to all the non-random local sampling areas.
7. The method for predicting the damage degree of rice leaf rollers according to claim 1, wherein the obtaining the number of the second disease leaves corresponding to the random local sampling area according to the number of the global total leaves corresponding to the random local sampling area and a preset one-time equation to obtain the total number of the second disease leaves corresponding to all random local sampling areas comprises:
inputting the total number of the global blades corresponding to each random local sampling area to a preset one-time equation corresponding to the random local sampling area, and obtaining the total number of the disease blades corresponding to each random local sampling area, which is output by the preset one-time equation;
And determining the total number of the second disease blades according to the sum of the total number of the disease blades corresponding to all the random local sampling areas.
8. The method for predicting the damage degree of rice leaf rollers according to claim 1, wherein determining the prediction result of rice diseases according to the total number of the first disease leaves, the total number of the second disease leaves and the total number of the leaves of each prediction subarea comprises:
determining the total number of the rice disease leaves in the area to be predicted according to the sum of the total number of the first disease leaves and the total number of the second disease leaves;
determining the total number of blades of the area to be predicted according to the sum of the total number of blades of the overall situation of each prediction subarea;
determining the leaf rolling rate of the rice diseases according to the total number of the rice disease leaves in the area to be predicted and the quotient of the total number of the leaves in the area to be predicted;
and determining the rice disease prediction result of the region to be predicted from a preset rice hazard degree table according to the rice disease leaf rolling rate.
9. The utility model provides a rice leaf roller harm degree prediction device which characterized in that includes:
the first determining unit is used for determining the total number of first disease blades corresponding to all the non-random local sampling areas according to the total global total blade number corresponding to all the non-random local sampling areas in the area to be predicted and a preset quadratic linear equation, and the area to be predicted comprises all the non-random local sampling areas and all the random local sampling areas;
The acquisition unit is used for acquiring the number of second disease blades corresponding to the random local sampling areas according to the number of global total blades corresponding to the random local sampling areas and a preset equation for each random local sampling area so as to acquire the total number of second disease blades corresponding to all the random local sampling areas;
the second determining unit is used for determining a rice disease prediction result according to the total number of the first disease blades, the total number of the second disease blades and the total number of the global total blades of each prediction subarea;
the preset quadratic equation is generated after fitting the number of the local disease blades and the number of the local total blades of all the random local sampling areas, and each preset quadratic equation is determined according to the number of the local disease blades and the number of the local total blades of each random local sampling area;
the number of the local disease blades and the number of the local total blades are determined by sampling at a first preset angle at any target position above the rice at a first preset distance from the rice;
the global total blade number is determined by sampling the center position above the rice in the area to be predicted at a second preset distance from the rice at a second preset angle; the first preset distance is smaller than the second preset distance, and the first preset angle is smaller than the second preset angle.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the rice leaf roller hazard level prediction method of any one of claims 1-8 when the program is executed.
11. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the rice leaf roller hazard level prediction method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310961079.3A CN117197655A (en) | 2023-08-01 | 2023-08-01 | Rice leaf roller hazard degree prediction method, device, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310961079.3A CN117197655A (en) | 2023-08-01 | 2023-08-01 | Rice leaf roller hazard degree prediction method, device, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117197655A true CN117197655A (en) | 2023-12-08 |
Family
ID=88982610
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310961079.3A Pending CN117197655A (en) | 2023-08-01 | 2023-08-01 | Rice leaf roller hazard degree prediction method, device, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117197655A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521830A (en) * | 2011-11-30 | 2012-06-27 | 浙江大学 | Optimum band selection method for hyperspectral images of canopy of crop under disease stress |
| CN108694391A (en) * | 2018-05-16 | 2018-10-23 | 黄铁成 | Populus Euphratica spring looper disaster monitoring method based on high-spectrum remote-sensing |
| CN111968003A (en) * | 2020-09-04 | 2020-11-20 | 郑州轻工业大学 | Crop disease prediction method based on crop ontology conceptual response |
| CN114399480A (en) * | 2021-12-30 | 2022-04-26 | 中国农业大学 | Method and device for detecting severity of vegetable leaf disease |
| JP2022094783A (en) * | 2020-12-15 | 2022-06-27 | 国立研究開発法人農業・食品産業技術総合研究機構 | Disease and insect damage diagnostic device, disease and insect damage diagnostic method, disease and insect damage diagnostic program, model generation device, model generation method, and model generation program |
| CN114925932A (en) * | 2022-06-13 | 2022-08-19 | 中国科学院空天信息创新研究院 | Crop disease severity prediction method and system, storage medium and electronic equipment |
| CN115035309A (en) * | 2022-04-29 | 2022-09-09 | 北京市农林科学院信息技术研究中心 | Rice disease identification method and device |
| CN116309366A (en) * | 2023-02-09 | 2023-06-23 | 北京市农林科学院智能装备技术研究中心 | Early-stage determination method and device for nematode disease of pine tree |
-
2023
- 2023-08-01 CN CN202310961079.3A patent/CN117197655A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521830A (en) * | 2011-11-30 | 2012-06-27 | 浙江大学 | Optimum band selection method for hyperspectral images of canopy of crop under disease stress |
| CN108694391A (en) * | 2018-05-16 | 2018-10-23 | 黄铁成 | Populus Euphratica spring looper disaster monitoring method based on high-spectrum remote-sensing |
| CN111968003A (en) * | 2020-09-04 | 2020-11-20 | 郑州轻工业大学 | Crop disease prediction method based on crop ontology conceptual response |
| JP2022094783A (en) * | 2020-12-15 | 2022-06-27 | 国立研究開発法人農業・食品産業技術総合研究機構 | Disease and insect damage diagnostic device, disease and insect damage diagnostic method, disease and insect damage diagnostic program, model generation device, model generation method, and model generation program |
| CN114399480A (en) * | 2021-12-30 | 2022-04-26 | 中国农业大学 | Method and device for detecting severity of vegetable leaf disease |
| CN115035309A (en) * | 2022-04-29 | 2022-09-09 | 北京市农林科学院信息技术研究中心 | Rice disease identification method and device |
| CN114925932A (en) * | 2022-06-13 | 2022-08-19 | 中国科学院空天信息创新研究院 | Crop disease severity prediction method and system, storage medium and electronic equipment |
| CN116309366A (en) * | 2023-02-09 | 2023-06-23 | 北京市农林科学院智能装备技术研究中心 | Early-stage determination method and device for nematode disease of pine tree |
Non-Patent Citations (2)
| Title |
|---|
| "Automatically diagnosing leaf scorching and disease symptoms in trees/shrubs by using RGB image computation with a common statistical algorithm", ECOLOGICAL INFORMATICS, vol. 38, 31 March 2017 (2017-03-31), pages 110 - 114 * |
| 檀根甲;李淼;江彤;孙俊铭;: "早稻品种对条纹叶枯病的田间抗性及病害侵染梯度", 植物保护, vol. 36, no. 5, 31 December 2010 (2010-12-31), pages 119 - 123 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111105011B (en) | CNN-based method and device for performing round-robin screening on useful learning data | |
| CN110120065B (en) | Target tracking method and system based on hierarchical convolution characteristics and scale self-adaptive kernel correlation filtering | |
| US20150154767A1 (en) | Method for reconstructing pet image using gpu parallel computing | |
| CN106600595A (en) | Human body characteristic dimension automatic measuring method based on artificial intelligence algorithm | |
| EP2851867A2 (en) | Method and apparatus for filtering an image | |
| CN114821102A (en) | Intensive citrus quantity detection method, equipment, storage medium and device | |
| CN114399480A (en) | Method and device for detecting severity of vegetable leaf disease | |
| CN113643365A (en) | Camera pose estimation method, device, equipment and readable storage medium | |
| CN112883915B (en) | Automatic wheat head identification method and system based on transfer learning | |
| CN113838087B (en) | Anti-occlusion target tracking method and system | |
| CN102314687B (en) | Method for detecting small targets in infrared sequence images | |
| CN117197655A (en) | Rice leaf roller hazard degree prediction method, device, electronic equipment and medium | |
| CN110472092B (en) | Geographical positioning method and system of street view picture | |
| CN112699836A (en) | Segmentation method and device for low-altitude paddy field image and electronic equipment | |
| CN111753775A (en) | Fish growth assessment method, device, equipment and storage medium | |
| Lara et al. | Weed estimation on lettuce crops using histograms of oriented gradients and multispectral images | |
| CN116246126A (en) | Iterative unsupervised domain self-adaption method and device | |
| CN110008881A (en) | The recognition methods of the milk cow behavior of multiple mobile object and device | |
| CN116232694A (en) | Lightweight network intrusion detection method and device, electronic equipment and storage medium | |
| CN116245901A (en) | Building rule vector contour extraction method and system based on edge learning | |
| CN115561480A (en) | Drainage outlet flow velocity detection method and system based on deep learning | |
| CN110796684B (en) | Target tracking method and related device | |
| CN111540418A (en) | Method and system for predicting probability value of excessive arsenic in plant | |
| CN111260681A (en) | Moving target tracking method and moving target tracking device | |
| US20110091106A1 (en) | Image Processing Method And System |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |