CN114359746A - Machine learning multispectral remote sensing image crop straw field-leaving extraction method and system - Google Patents
Machine learning multispectral remote sensing image crop straw field-leaving extraction method and system Download PDFInfo
- Publication number
- CN114359746A CN114359746A CN202111634631.5A CN202111634631A CN114359746A CN 114359746 A CN114359746 A CN 114359746A CN 202111634631 A CN202111634631 A CN 202111634631A CN 114359746 A CN114359746 A CN 114359746A
- Authority
- CN
- China
- Prior art keywords
- crop
- field
- sample set
- remote sensing
- field block
- 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
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The multispectral remote sensing image crop straw field leaving extraction method and system based on machine learning solve the problems that human errors and measurement errors exist in field leaving area and progress of an existing straw field leaving monitoring mode, and accurate management and control cannot be achieved for positioning of a field leaving area. The method comprises the following steps: step S1, acquiring multispectral image data of a crop growth period, sketching and establishing a polygonal sample set according to crop spectral information, establishing a crop classification model, and predicting the established crop classification model to obtain a crop distribution result; step S2, acquiring multispectral image data after a crop harvesting period, sketching polygonal samples of corresponding categories, and then performing sampling processing to obtain a field block-separated polygonal sample set; and step S3, dividing the field block polygonal sample set by utilizing five-fold division, obtaining five groups of training sets-verification sets which are trained to be five base models, predicting image data by using the base models, and determining the field block by fusing the prediction results through probability mean values.
Description
Technical Field
The invention relates to the technical field of multispectral satellite remote sensing image data mining, in particular to a method and a system for extracting field blocks from multispectral remote sensing images in machine learning.
Background
With the rapid development of agriculture in China, the straw treatment of crops gradually becomes an important part of agriculture. At present, the treatment mode of the straws is mainly a mode of separating the straws from the field, and the straw separation mode mainly comprises two modes, namely, the straws are cut or crushed and uniformly spread on the ground surface, and then are deeply ploughed into the soil by a high-power tractor through ploughing or rotary tillage; the other is a mode of directly packing and transporting away the straws. The air pollution caused by direct burning of the straws can be reduced and the straws can be recycled by leaving the field, so that the method has important significance for agricultural environment protection and promotion of yield increase and income increase of farmers.
At present, agricultural and environmental protection related departments also supervise and manage farmland straw leaving from the field in various ways, and the supervision and field leaving mode effectively improves the utilization rate and the leaving rate of the straw and reduces the phenomenon of illegal burning of the straw by supervising and urging the straw leaving from the field. However, the main mode for acquiring the straw field leaving information at present is to count and report by farmers or cooperative society, and the main defects of the mode are as follows:
1) the area and the progress of the straw leaving the field can not be accurately measured, and certain personal errors and measurement errors exist.
2) The area which is not away from the field cannot be positioned accurately.
To sum up, although the existing straw field leaving monitoring mode can effectively improve the utilization rate of straws, the field leaving rate and reduce the phenomenon of straw illegal burning, certain personal errors and measuring errors exist in the monitoring of the area and the progress of the straws leaving the field, and accurate management and control cannot be realized in the positioning of the area which is not leaving the field.
Disclosure of Invention
The method solves the problems that the existing method for monitoring the field leaving area and progress of the straw field leaving mode has human errors and measurement errors, and accurate management and control cannot be realized on the positioning of the area which is not separated from the field.
The invention relates to a machine-learning multispectral remote sensing image crop straw field block extraction method, which comprises the following steps:
step S1, acquiring multispectral image data of a crop growth period, sketching and establishing a polygonal sample set according to crop spectral information, establishing a crop classification model, and predicting the established crop classification model to obtain a crop distribution result;
step S2, acquiring multispectral image data after a crop harvesting period, sketching polygonal samples of corresponding categories, and then performing sampling processing to obtain a field block-separated polygonal sample set;
and step S3, dividing the field block polygonal sample set by utilizing five-fold division, obtaining five groups of training sets-verification sets which are trained to be five base models, predicting image data by using the base models, and determining the field block by fusing the prediction results through probability mean values.
Further, in an embodiment of the present invention, in the acquiring the multispectral image data of the growing period of the crop at step S1, the multispectral image data refers to a wavelength band of:
b2, B3, B4, B5, B6, B7, B8, B8A, B11, B12, and the multispectral image data is in the form of 16 bits without symbol shaping.
Further, in an embodiment of the present invention, the step S1 is to delineate and establish a polygon sample set according to the crop spectrum information,
the crop extraction comprises the following steps: corn, rice, other crops and other ground and object crops,
and drawing and establishing a polygon sample set, namely drawing and establishing 50-100 polygon samples for each type of crops to form the sample set.
Further, in an embodiment of the present invention, the establishing of the crop classification model in the step S1 is based on the LightGBM framework.
Further, in an embodiment of the present invention, in the polygon samples that delineate corresponding categories in step S2, the categories include:
leaving field, not leaving field and other ground features.
Further, in an embodiment of the present invention, the performing the sampling processing in step S2 includes the following steps:
step S201, the wave band value of each pixel in a polygonal sample obtained by calculation is used as a basic wave band characteristic value, and a corresponding category spectral index characteristic value is calculated based on the wave band characteristic value;
step S202, normalization processing is carried out on the spectral index characteristic value.
Further, in an embodiment of the present invention, the method for dividing the field block polygonal sample set by using five-fold division in step S3 is:
the set of outlier polygon samples is scaled by 4: the method 1 is divided into a training set and a verification set.
Further, in an embodiment of the present invention, the prediction results of the step S3 are merged by means of probability mean as follows:
and recombining the block probability matrixes obtained by predicting the five basic models, adding the probability matrixes and calculating the mean value to obtain a total probability matrix after fusion.
The invention relates to a machine-learned multispectral remote sensing image crop straw field block extraction system, which comprises the following modules:
the module S1 is used for acquiring multispectral image data of a crop growth period, sketching and establishing a polygonal sample set according to crop spectral information, establishing a crop classification model for the polygonal sample set, and predicting the established crop classification model to obtain a crop distribution result;
the module S2 is used for acquiring multispectral image data after the crop harvest period, sketching polygon samples of corresponding categories, and then performing sampling processing to obtain a polygon sample set away from a field block;
and the module S3 is used for dividing the field block polygonal sample set by utilizing five-fold division to obtain five groups of training sets, namely the verification set, which are trained to form five base models, and is also used for predicting image data by using the base models, and determining the field block by fusing the prediction result through probability mean.
A computer-readable storage medium according to the invention, on which a computer program is stored, is characterized in that the computer program, when being executed by a processor, carries out the steps of the method as claimed in any one of the above-mentioned methods.
A computer device according to the present invention comprises a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that the processor executes the steps of the method according to any one of the above methods when executing the computer program stored in the memory.
The method solves the problems that the existing method for monitoring the field leaving area and progress of the straw field leaving mode has human errors and measurement errors, and accurate management and control cannot be realized on the positioning of the area which is not separated from the field. The method has the following specific beneficial effects:
1. the invention provides a new way for monitoring the straw field leaving mode by monitoring the field leaving and field non-leaving blocks of the remote sensing satellite image for the first time.
2. According to the method, the field leaving and field not leaving blocks in the remote sensing satellite image are extracted, so that important data support is provided for monitoring the field leaving and field not leaving areas and the field leaving progress, and the problem of measurement errors of the field leaving areas and the field leaving progress in the conventional straw field leaving monitoring mode is solved.
3. By extracting field leaving and field non-leaving blocks in the remote sensing satellite image, the method not only realizes accurate monitoring of the field leaving and field non-leaving areas and reduces the cost of the existing manual monitoring mode, but also effectively monitors the field leaving progress of the straw and prevents illegal burning of the straw, and has important significance for efficient utilization of agricultural resources and environmental protection.
4. The method aims at extracting spectral features and index features of multispectral satellite remote sensing images, and establishes a straw field-leaving extraction model in a machine learning mode, so that the problem that field-leaving blocks and field-not-leaving blocks cannot be distinguished is solved.
The method is suitable for the technical field of remote sensing field leaving monitoring in a large area, automatic extraction is realized by establishing a straw field leaving detection model, the technical problem of low straw field leaving interpretation efficiency in the large area range is solved, the manual participation amount is greatly reduced, and the extraction of the straw field leaving can be quickly carried out after the data acquisition process to realize effective control of the field leaving progress.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flow chart of the construction of the multispectral remote sensing image crop straw field block extraction method based on machine learning.
Fig. 2 is a band synthesis color chart according to the fifth embodiment.
Fig. 3 is a diagram of the prediction result of the straw field block classification model according to the tenth embodiment, (a) a color map of an original image 1, (b) a diagram of an extraction result of an original image 1 from the field, (c) a color map of an original image 2, and (d) a diagram of an extraction result of an original image 2 from the field.
Detailed Description
Various embodiments of the present invention will now be described more fully hereinafter with reference to the accompanying drawings. The embodiments described by referring to the drawings are exemplary and intended to be illustrative of the invention and are not to be construed as limiting the invention.
In a first embodiment, a method for extracting a crop straw field block from a multispectral remote sensing image by machine learning according to the first embodiment can be better understood with reference to fig. 1, and includes the following steps:
step S1, acquiring multispectral image data of a crop growth period, sketching and establishing a polygonal sample set according to crop spectral information, establishing a crop classification model, and predicting the established crop classification model to obtain a crop distribution result;
step S2, acquiring multispectral image data after a crop harvesting period, sketching polygonal samples of corresponding categories, and then performing sampling processing to obtain a field block-separated polygonal sample set;
and step S3, dividing the field block polygonal sample set by utilizing five-fold division, obtaining five groups of training sets-verification sets which are trained to be five base models, predicting image data by using the base models, and determining the field block by fusing the prediction results through probability mean values.
In the embodiment, the main straw-producing crops in the cultivated land are extracted, and corn and rice are extracted as two main straw-producing crops.
The growth period of the corn and rice straws is 7-8 months, the corn and rice harvesting period is 10-11 months later, the corn and rice in 10-11 months are harvested, and the collection of corresponding multispectral image data according to the growth periods is more beneficial to the extraction of the non-field-leaving area.
In a second embodiment, in the method for extracting a crop straw field block from a machine-learned multispectral remote sensing image according to the first embodiment, in the present embodiment, in the step S1, the multispectral image data obtained in the growing period of the crop is a multispectral image having a wavelength band of:
b2, B3, B4, B5, B6, B7, B8, B8A, B11, B12, and the multispectral image data is in the form of 16 bits without symbol shaping.
In this embodiment, the multispectral image selects the reflectivity products of 10 bands because the spectral characteristics of the field-leaving and field-non-leaving have a certain difference in the band, and the machine learning algorithm used in the method has a better effect on multiple features, and can automatically reduce the weight of the features that do not improve the model accuracy. The accuracy of the whole straw field-leaving prediction can be improved by means of feature extraction and wave band combination on the 10 wave bands.
In a third embodiment, in the method for extracting a block from a field of crop straw by using a machine-learned multispectral remote sensing image according to the first embodiment, in the present embodiment, the polygon sample set is sketched and established according to the crop spectrum information in the step S1,
the crop extraction comprises the following steps: corn, rice, other crops and other ground and object crops,
and drawing and establishing a polygon sample set, namely drawing and establishing 50-100 polygon samples for each type of crops to form the sample set.
In this embodiment, the band feature value of each pixel is extracted from the outlined sample polygon as a training set.
In a fourth embodiment, the present invention is a method for extracting a crop straw off-field block from a multispectral remote sensing image obtained by machine learning according to the first embodiment, wherein the step S1 of establishing a crop classification model is based on a LightGBM framework.
In the embodiment, a crop classification model is established based on the LightGBM framework, and model prediction is performed on a satellite image of a target area by using the model to obtain a corn and rice crop distribution result of the target area to be extracted, namely an original straw distribution result. In the subsequent step, the crop distribution result is used for masking the field leaving extraction result, so that the accuracy of the subsequent straw field leaving extraction result can be ensured.
In a fifth embodiment, the present embodiment is a method for extracting a crop straw field block from a multispectral remote sensing image by machine learning according to the first embodiment, in the present embodiment, the polygon samples delineating corresponding categories in step S2 are:
leaving field, not leaving field and other ground features.
In this embodiment, "other ground objects" need to be drawn to include water, impervious surface, forest land and other first-class classes, so as to eliminate the influence of irrelevant ground objects in the classification model.
Specifically, in order to enhance the difference between field-leaving straw and non-field-leaving straw and make the delineation result more accurate, a true color image is reestablished by using a wave band combination mode. The specific method comprises the following steps: the band synthesis graph shown in fig. 2 was obtained by performing stretching processing using B12 minus the B11 band as the red band, B9 as the green band, and B10 as the blue band, with n being 2 standard deviation. In the figure, the bright yellow areas are areas not leaving the field, and the dark yellow areas are areas leaving the field to different degrees. The wave band composite map can be used for clearly distinguishing land parcels which are separated from fields and land parcels which are not separated from fields.
In a sixth embodiment, the present invention is a method for extracting a crop straw field block from a multispectral remote sensing image by machine learning according to the first embodiment, wherein the sampling processing in step S2 includes the steps of:
step S201, the wave band value of each pixel in a polygonal sample obtained by calculation is used as a basic wave band characteristic value, and a corresponding category spectral index characteristic value is calculated based on the wave band characteristic value;
step S202, normalization processing is carried out on the spectral index characteristic value.
In this embodiment, each polygon sample in the vector data needs to be read first, the image coordinates of all pixels in each sample polygon are obtained, the band value of each pixel in one sample is calculated and used as the fundamental band feature, and then various spectral index features are calculated based on the band feature.
1) Normalized farming index NDTI
2) Normalized straw index NDRI
3) Simple farming index STI
4) Normalized difference index NDI7
In the formula, B11 and B12 represent short-wave infrared bands which are sensitive to the distance from the field. B4 and B8 respectively represent the red band and the near-red band. The accuracy of the model is effectively improved by adding the following spectral index features. And combining the wave band characteristics with the spectral index model to obtain a multi-characteristic sample data set.
In order to ensure the generalization of the model, normalization processing needs to be performed on the feature values. The band features are normalized to be between 0 and 1 by using a mode of maximum and minimum values, and the normalization formula is as follows:
in the formula, Xmax、XminRespectively representing the maximum and minimum of the spectral values of the region, X representing each spectral value, XnewRepresenting normalized spectral values. Since the spectral index features themselves are in the form of ratios, no normalization calculation is required. Based on the above steps, a sample set containing 14 features of multiple classes can be obtained.
A seventh embodiment is a method for extracting a crop straw field block from a machine-learned multispectral remote sensing image according to the first embodiment, wherein in the present embodiment, the method for dividing a field block polygonal sample set by a quintuple method in step S3 includes:
the set of outlier polygon samples is scaled by 4: the method 1 is divided into a training set and a verification set.
In the embodiment, in order to fully utilize all data, the data set is manufactured by using a five-fold division mode, and all data sets can be ensured to participate in the model training process by using the method.
In an eighth embodiment, the present embodiment is a method for extracting a crop straw field block from a multispectral remote sensing image by machine learning according to the seventh embodiment, wherein the method for obtaining five training sets-verification set training in step S3 includes:
and taking five training sets and verification sets, wherein 4 of the training sets and 1 of the verification sets are taken as the training sets and the verification sets at each time, so as to obtain five groups of training sets-verification sets.
In a ninth embodiment, the present embodiment is a method for extracting a crop straw field block from a multispectral remote sensing image by machine learning according to the seventh embodiment, wherein the method for acquiring five basis models in step S3 includes:
and respectively establishing classification models for the five groups of training sets-verification sets by adopting a LightGBM framework, wherein the model training turn is 150, the learning rate is 0.1, the characteristic sampling probability is 50%, and the sample sampling probability is 50%.
In this embodiment, after the training is completed, five basis models can be obtained. After model training is completed, the image data is predicted by using five base models respectively. Because the global image is large and cannot be directly read into the memory for processing, a block prediction mode is required to reduce the memory consumption. Firstly, after an original remote sensing image is read, generally selecting the block size as 4096 and the overlapping rate as 10% according to the preset block size and the block overlapping rate, generating a small-size image from remote sensing image data in the blocks, and flattening a data set into a single-pixel single sample. Then inputting the block probability matrix into a model and predicting to obtain the probability matrix of each block.
In a tenth embodiment, in the present embodiment, the method for extracting a block from a crop straw field by using a machine-learned multispectral remote sensing image according to the first embodiment is a method for extracting a block from a crop straw field by using machine learning, wherein the prediction results in the step S3 are integrated into a set of values by using a probability mean:
and recombining the block probability matrixes obtained by predicting the five basic models, adding the probability matrixes and calculating the mean value to obtain a total probability matrix after fusion.
In this embodiment, the block probability matrices predicted by the five models are recombined, the probability matrices are added and an average value is calculated to obtain a total probability matrix after fusion, the category value of each pixel is determined according to the maximum index of the probability matrix, and the field-leaving block of the target area is determined according to the category. Masking the extracted corn and rice crop results to remove the pattern spots of the non-corn and rice crop regions to obtain more accurate field-leaving plot results, as shown in fig. 3, which are the extraction results of two sample images, wherein (a) an original image 1 color map, (b) an original image 1 field-leaving extraction result map, (c) an original image 2 color map, and (d) an original image 2 field-leaving extraction result map.
(b) In the graph of (d), the black areas represent the land blocks extracted by the model, and the white areas represent the non-land blocks and other non-land areas.
Selecting a proper projection coordinate system according to the position of the target area, calculating the area of the field-leaving plot, and obtaining the field-leaving rate of each village and town according to a field-leaving rate calculation formula, wherein the field-leaving rate is calculated as follows:
eleventh, the system for extracting crop straw from field blocks by using machine-learned multispectral remote sensing images according to the present embodiment includes the following modules:
the module S1 is used for acquiring multispectral image data of a crop growth period, sketching and establishing a polygonal sample set according to crop spectral information, establishing a crop classification model for the polygonal sample set, and predicting the established crop classification model to obtain a crop distribution result;
the module S2 is used for acquiring multispectral image data after the crop harvest period, sketching polygon samples of corresponding categories, and then performing sampling processing to obtain a polygon sample set away from a field block;
and the module S3 is used for dividing the field block polygonal sample set by utilizing five-fold division to obtain five groups of training sets, namely the verification set, which are trained to form five base models, and is also used for predicting image data by using the base models, and determining the field block by fusing the prediction result through probability mean.
Twelfth, a computer-readable storage medium according to this embodiment, wherein a computer program is stored thereon, and when being executed by a processor, the computer program implements the steps of the method according to any one of the first to tenth embodiments.
Embodiment thirteen is a computer device including a memory and a processor, the memory storing a computer program, and in this embodiment, when the processor executes the computer program stored in the memory, the method of any one of embodiment one to ten is performed.
The above embodiments are examples of the method for extracting crop straw field-away blocks based on machine learning multispectral remote sensing images, and the protection scope of the present invention also includes reasonable combinations of features defined in the above embodiments.
Claims (11)
1. The machine-learned multispectral remote sensing image crop straw field block extraction method is characterized by comprising the following steps of:
step S1, acquiring multispectral image data of a crop growth period, sketching and establishing a polygonal sample set according to crop spectral information, establishing a crop classification model, and predicting the established crop classification model to obtain a crop distribution result;
step S2, acquiring multispectral image data after a crop harvesting period, sketching polygonal samples of corresponding categories, and then performing sampling processing to obtain a field block-separated polygonal sample set;
and step S3, dividing the field block polygonal sample set by utilizing five-fold division, obtaining five groups of training sets-verification sets which are trained to be five base models, predicting image data by using the base models, and determining the field block by fusing the prediction results through probability mean values.
2. The machine-learned multispectral remote sensing image crop straw field block extraction method according to claim 1, wherein in the step S1 of acquiring multispectral image data of a growing period of a crop, the multispectral image data refer to a wave band:
b2, B3, B4, B5, B6, B7, B8, B8A, B11, B12, and the multispectral image data is in the form of 16 bits without symbol shaping.
3. The method for machine-learned multispectral remote sensing image crop straw field block extraction according to claim 1, wherein the step S1 is performed by sketching and creating a polygonal sample set according to crop spectrum information,
the crop extraction comprises the following steps: corn, rice, other crops and other ground and object crops,
and drawing and establishing a polygon sample set, namely drawing and establishing 50-100 polygon samples for each type of crops to form the sample set.
4. The machine-learned multispectral remote sensing image crop straw field-off block extraction method as claimed in claim 1, wherein the establishing of the crop classification model in step S1 is based on a LightGBM framework.
5. The machine-learned multispectral remote sensing image crop straw field block extraction method according to claim 1, wherein the polygon samples of step S2 that delineate corresponding categories include:
leaving field, not leaving field and other ground features.
6. The method for extracting crop straw field block from machine-learned multispectral remote sensing image according to claim 1, wherein the sampling process in step S2 includes the following steps:
step S201, the wave band value of each pixel in a polygonal sample obtained by calculation is used as a basic wave band characteristic value, and a corresponding category spectral index characteristic value is calculated based on the wave band characteristic value;
step S202, normalization processing is carried out on the spectral index characteristic value.
7. The machine-learned multispectral remote sensing image crop straw field block extraction method according to claim 1, wherein the method for dividing the field block polygonal sample set by using quintuple-fold division in the step S3 is as follows:
the set of outlier polygon samples is scaled by 4: the method 1 is divided into a training set and a verification set.
8. The machine-learned multispectral remote sensing image crop straw field block extraction method as claimed in claim 1, wherein the prediction results of the step S3 are merged into by means of probability mean values:
and recombining the block probability matrixes obtained by predicting the five basic models, adding the probability matrixes and calculating the mean value to obtain a total probability matrix after fusion.
9. Multispectral remote sensing image crop straw off-field block extraction system of machine learning, its characterized in that, the system includes following module:
the module S1 is used for acquiring multispectral image data of a crop growth period, sketching and establishing a polygonal sample set according to crop spectral information, establishing a crop classification model for the polygonal sample set, and predicting the established crop classification model to obtain a crop distribution result;
the module S2 is used for acquiring multispectral image data after the crop harvest period, sketching polygon samples of corresponding categories, and then performing sampling processing to obtain a polygon sample set away from a field block;
and the module S3 is used for dividing the field block polygonal sample set by utilizing five-fold division to obtain five groups of training sets, namely the verification set, which are trained to form five base models, and is also used for predicting image data by using the base models, and determining the field block by fusing the prediction result through probability mean.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 8.
11. A computer device comprising a memory and a processor, the memory having a computer program stored therein, characterized in that the steps of the method of any of claims 1 to 8 are performed when the processor runs the computer program stored by the memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111634631.5A CN114359746A (en) | 2021-12-29 | 2021-12-29 | Machine learning multispectral remote sensing image crop straw field-leaving extraction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111634631.5A CN114359746A (en) | 2021-12-29 | 2021-12-29 | Machine learning multispectral remote sensing image crop straw field-leaving extraction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114359746A true CN114359746A (en) | 2022-04-15 |
Family
ID=81103313
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111634631.5A Pending CN114359746A (en) | 2021-12-29 | 2021-12-29 | Machine learning multispectral remote sensing image crop straw field-leaving extraction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114359746A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115761518A (en) * | 2023-01-10 | 2023-03-07 | 云南瀚哲科技有限公司 | Crop classification method based on remote sensing image data |
| CN116129280A (en) * | 2023-04-17 | 2023-05-16 | 北京数慧时空信息技术有限公司 | Method for detecting snow in remote sensing image |
| CN117789025A (en) * | 2023-12-26 | 2024-03-29 | 北京工业大学 | Open-air incineration emission estimation method for crop straws based on climate information |
| CN117789025B (en) * | 2023-12-26 | 2024-06-04 | 北京工业大学 | Open-air incineration emission estimation method for crop straws based on climate information |
-
2021
- 2021-12-29 CN CN202111634631.5A patent/CN114359746A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115761518A (en) * | 2023-01-10 | 2023-03-07 | 云南瀚哲科技有限公司 | Crop classification method based on remote sensing image data |
| CN115761518B (en) * | 2023-01-10 | 2023-04-11 | 云南瀚哲科技有限公司 | Crop classification method based on remote sensing image data |
| CN116129280A (en) * | 2023-04-17 | 2023-05-16 | 北京数慧时空信息技术有限公司 | Method for detecting snow in remote sensing image |
| CN117789025A (en) * | 2023-12-26 | 2024-03-29 | 北京工业大学 | Open-air incineration emission estimation method for crop straws based on climate information |
| CN117789025B (en) * | 2023-12-26 | 2024-06-04 | 北京工业大学 | Open-air incineration emission estimation method for crop straws based on climate information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20230292647A1 (en) | System and Method for Crop Monitoring | |
| Qi et al. | Monitoring of peanut leaves chlorophyll content based on drone-based multispectral image feature extraction | |
| Kawamura et al. | Discriminating crops/weeds in an upland rice field from UAV images with the SLIC-RF algorithm | |
| CN110909679B (en) | Remote sensing identification method and system for fallow crop rotation information of winter wheat historical planting area | |
| CN114359746A (en) | Machine learning multispectral remote sensing image crop straw field-leaving extraction method and system | |
| Zaman et al. | Estimation of wild blueberry fruit yield using digital color photography | |
| Rasti et al. | A survey of high resolution image processing techniques for cereal crop growth monitoring | |
| Chen et al. | Predicting individual apple tree yield using UAV multi-source remote sensing data and ensemble learning | |
| Zhang et al. | EPSA-YOLO-V5s: A novel method for detecting the survival rate of rapeseed in a plant factory based on multiple guarantee mechanisms | |
| Lootens et al. | High-throughput phenotyping of lateral expansion and regrowth of spaced Lolium perenne plants using on-field image analysis | |
| CN113780097A (en) | Arable land extraction method based on knowledge map and deep learning | |
| CN110321774A (en) | Crops evaluation methods for disaster condition, device, equipment and computer readable storage medium | |
| Lee et al. | Single-plant broccoli growth monitoring using deep learning with UAV imagery | |
| Li et al. | Estimation of nitrogen content in wheat using indices derived from RGB and thermal infrared imaging | |
| CN108537164B (en) | Method and device for monitoring germination rate of dibbling and sowing based on unmanned aerial vehicle remote sensing | |
| Latha et al. | Technology for kisan samanvayam: Nutrition intelligibility of groundnut plant using IoT-ML framework | |
| Zhu et al. | UAV flight height impacts on wheat biomass estimation via machine and deep learning | |
| Ormeci et al. | Identification of crop areas Using SPOT–5 data | |
| Dobbs et al. | Using structure-from-motion to estimate cover crop biomass and characterize canopy structure | |
| Duman et al. | Design of a smart vertical farming system using image processing | |
| Yang et al. | Feature extraction of cotton plant height based on DSM difference method | |
| Zou et al. | Combining spectral and texture feature of UAV image with plant height to improve LAI estimation of winter wheat at jointing stage | |
| Mangla et al. | Statistical growth prediction analysis of rice crop with pixel-based mapping technique | |
| CN113793376B (en) | Irrigation water body extraction method | |
| CN116453003B (en) | Method and system for intelligently identifying rice growth vigor based on unmanned aerial vehicle monitoring |
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 |