30 results about "Crop phenology" patented technology

The invention discloses a method and a system for monitoring crop growth, which relates to the technical field of agricultural monitoring. The method comprises the following steps that a growth period of crops in a monitoring area is obtained; remote sensing data of the current year and historical remote sensing data in the time range of the growth period is obtained and pretreated to obtain vegetation index data of each pixel of the monitoring area; maximum and minimum values synthesis is carried out in the time range of the growth period to obtain the maximum value of the vegetation index in the growth period of the current year and the maximum and the minimum values of the vegetation index in the growth period of the historical year; the optimal vegetation index in the growth period of the monitoring area is obtained by calculation, and the situation of the current crop growth is determined according to the value and the level threshold range of the optimal vegetation index. The influence of unfavorable weather conditions on the vegetation index during the growth period of the crop is effectively reduced and the effect of the phenological migration on the monitoring result of the crop growth is overcome.

The invention discloses a sophisticated crop classifying method based on combination of object oriented technology and NDVI (normalized difference vegetation index) time series. The characteristic that the NDVI time series can reflect crop phenology information is made into full use for realizing crop distinguishing. In addition, the salt-and-pepper effect can be eliminated on the basis of the object oriented technology, precision in sophisticated crop classification is effectively improved, and a new conception is provided for sophisticated crop classification.

The invention discloses a multi-time sequence image rice yield estimation method based on a crop phenological period, belongs to the technical field of remote sensing technology and crop yield estimation, and aims to solve the problems that the precision of an inversion result is inaccurate, a model is unstable, and the influence of growth vigor, disasters and soil moisture content is not considered. According to the method, a BP neural network algorithm is adopted, matlab2019b programming and GUI design are applied to establish a rice yield estimation model, model training is performed by applying a vegetation index of a rice phenological period, rice growth trend grade data, rice disaster grade data, soil moisture content data and field yield estimation data, an optimal model is established, the optimal model is combined with a target area image to perform rice yield estimation, and the output grade distribution grid number of the target area is obtained. The remote sensing technology is used for processing satellite image data and extracting vegetation indexes, growth vigor, disasters and soil moisture content, the yield of a target area is inversed, the production cost is saved, the production efficiency is improved, and meanwhile the purpose of accurately estimating the yield is achieved.

The invention provides an agriculture and forestry area returned farmland information collection method, which comprises the steps of collecting an NDVI sequence value within one year according to a remote sensing image with low space and high time resolution; calculating a CV value of any year and extracting crop phenological information according to the NDVI sequence value, and further determining planting information of any pixel area; according to the planting information and the crop phenology information, determining a multiple cropping index of any pixel area within a monitoring year; and based on the multiple cropping index and in combination with a remote sensing image sequence with medium-high spatial resolution, accurately determining the farmland returning time, the farmland returning area and the farmland returning position. According to the method, the NDVI value of the time sequence is collected by utilizing the agricultural and forestry area remote sensing image with low space and high time resolution, the returned farmland position, returned farmland time and returned farmland area are calculated, then the returned farmland information is collected by applying theremote sensing image with medium-high resolution and combining different multiple cropping index remote sensing models, and the information extraction precision is improved.

The invention discloses a crop intelligent remote sensing extraction method and system based on transfer learning. Based on geospatial autocorrelation, a partition transfer learning strategy is adopted, a target image is partitioned in a geographic subarea mode, it is ensured that a training sample of a machine learning model of an image of each geographic subarea comes from a classification result of an adjacent image, classification learning is carried out in a small range, and the classification result of the machine learning model of the image of each geographic subarea is obtained. Classification learning results are gradually extrapolated to adjacent sub-regions, crops in a research region are extracted from the local part to the whole part, phenological characteristic differences, caused by regional differences, of the same kind of crops can be effectively controlled, and classification precision is improved.

The invention discloses an insect pest early warning system based on the Internet of Things. The insect pest early warning system comprises a data acquisition subsystem, a data storage subsystem, a data analysis subsystem and an insect pest early warning subsystem, the data acquisition subsystem comprises an environment information acquisition module and an image acquisition module; the data storage subsystem comprises an insect pest basic database, a crop characteristic database and an environment information database; and the data analysis subsystem is used for processing the data acquired by the data acquisition subsystem through the crop phenological period prediction model and the insect pest effective accumulated temperature prediction model in combination with the insect pest basic database and the prediction characteristics of each model, and outputting the processed data to the insect pest early warning subsystem. According to the method, the occurrence date of each insect state of different insect pests is predicted by combining the characteristics of the crop phenological period prediction model and the insect pest effective accumulated temperature prediction model, the prediction is accurate and effective, the pre-prevention work of crops is facilitated, the operation is simple, and the method has a relatively high application value.

The invention discloses a crop leaf area index remote sensing inversion method under the constraint of land parcel spatial-temporal characteristics, and the method comprises the steps: simulating canopy directional reflection under the conditions of different leaf physical and chemical parameters, canopy structure parameters and soil properties based on a PROSAIL radiation transmission model, and constructing a lookup table of a land parcel; based on the normalized vegetation index time sequence data, extracting growth starting date features of crops in the land parcels; the crop leaf area index priori knowledge is constructed by combining the growth starting date characteristics of the crops in the land parcels and the MODIS leaf area index product; and then obtaining a lookup table subset of the land parcel, and performing inversion on the leaf area index of the land parcel through a cost function about crop parameter spatial autocorrelation to obtain an inversion result. According to the method, a new cost function is designed from the characteristic that the leaf area index of the crop changes along with the phenology of the crop and the spatial autocorrelation of the crop parameters in the land parcel, the leaf area index is inverted through the lookup table method, the effect is better, and the obtained result is higher in precision.

The invention provides a wheat phenological period real-time classification method based on an unmanned aerial vehicle RGB image, and the method comprises the steps: (1) obtaining a time sequence high-spatial-resolution RGB image according to the actual growth condition of a wheat field processed at different sowing periods, and carrying out the preprocessing of the image, and obtaining the unmanned aerial vehicle image of the same region in different years; (2) extracting spectral information and texture information of the time sequence unmanned aerial vehicle image, and deriving all spectral features and texture features as a feature complete set; (3) sorting all feature importance by a feature selection algorithm based on a compact-separation principle, and determining an optimal feature and a feature number; and (4) automatically classifying and identifying the features of different phenological stages by applying an mRVM classifier to obtain the overall classification precision and the classification precision of each period. The classification method constructed by the invention is simple and efficient, timely crop phenological information can be obtained, and a basis is provided for effectively guiding agricultural management decisions, such as irrigation, fertilization and pesticide management activities at a specific stage.

The invention relates to the technical field of agriculture, in particular to a development period simulation method based on a response and adaptation mechanism of crops to the environment. The method comprises the following steps: collecting actually measured crop phenology data of a research site; obtaining the number of days required by the development stage, the initial date sequence of the development stage and the average temperature of the development stage according to the phenological data; obtaining a development rate according to the required number of days of the development stage, wherein the development rate is the reciprocal of the required number of days of the development stage; calculating the value of the average temperature * the starting day sequence; obtaining a regression equation by taking the development rate as a dependent variable and taking the average temperature, the average temperature and an initial day sequence as independent variables through a binary one-time regression method, wherein the equation is a development period simulation formula. According to the method, a mechanism that crops respond to environment changes is considered, and a mechanism that the crops adapt to the environment is also considered, so that the development period can be better simulated compared with an existing model. The method is simple, only three parameters are needed, and the parameters can be directly obtained through observation data.

The invention discloses a crop distribution map, which is used for calculating the area of each type of crops on each grid according to grid-scale crop planting mode information extracted from remotesensing data and the area proportion of each type of crops on the planting mode of a corresponding administrative unit so as to make a crop spatial distribution map. According to the invention, the crop phenological characteristics of remote sensing recognition are combined with the actual phenological characteristics of the crops to determine the spatial distribution condition of the crops, and large-range, multi-scale and high-spatial heterogeneity crop spatial distribution drawing can be realized through remote sensing macroscopic, rapid acquisition, multi-scale and other characteristics.

The invention discloses a crop area and phenological index extraction method, which comprises the following steps of: preprocessing an SAR (Synthetic Aperture Radar) image, including track correction, thermal noise removal, radiometric calibration, speckle filtering, terrain correction and decibel processing, so as to obtain a backscattering coefficient of crop VV/VH; constructing a feature vector according to a backscattering coefficient of crop VV/VH, proposing a full convolutional neural network model based on a time sequence, and meanwhile, performing crop planting area drawing according to time sequence SAR (Synthetic Aperture Radar) image data; in combination with an SAR-VV/VH method, a crop phenological index is extracted according to a regional crop growth cycle; compared with the existing random forest algorithm, the extraction method based on the full convolutional neural network has the advantages that the crop extraction precision is high, the spatial texture information of the crop plot can be relatively completely reserved, the purpose of accurate crop drawing is achieved, and the method has high applicability in crop extraction.