CN117079152A - Fine crop classification extraction method and system based on satellite remote sensing image - Google Patents
Fine crop classification extraction method and system based on satellite remote sensing image Download PDFInfo
- Publication number
- CN117079152A CN117079152A CN202310847822.2A CN202310847822A CN117079152A CN 117079152 A CN117079152 A CN 117079152A CN 202310847822 A CN202310847822 A CN 202310847822A CN 117079152 A CN117079152 A CN 117079152A
- Authority
- CN
- China
- Prior art keywords
- classification
- remote sensing
- extraction
- crops
- time
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 127
- 229920000742 Cotton Polymers 0.000 claims abstract description 78
- 244000068988 Glycine max Species 0.000 claims abstract description 76
- 235000010469 Glycine max Nutrition 0.000 claims abstract description 76
- 238000012706 support-vector machine Methods 0.000 claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 38
- 238000001228 spectrum Methods 0.000 claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 238000005457 optimization Methods 0.000 claims abstract description 6
- 230000003595 spectral effect Effects 0.000 claims description 34
- 238000013145 classification model Methods 0.000 claims description 33
- 230000012010 growth Effects 0.000 claims description 20
- 241000209094 Oryza Species 0.000 claims description 16
- 235000007164 Oryza sativa Nutrition 0.000 claims description 16
- 235000009566 rice Nutrition 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 15
- 238000003066 decision tree Methods 0.000 claims description 15
- 241000219146 Gossypium Species 0.000 claims description 14
- 238000007635 classification algorithm Methods 0.000 claims description 14
- 238000011835 investigation Methods 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 10
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 5
- 238000002310 reflectometry Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 5
- 238000007621 cluster analysis Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 abstract description 6
- 238000010183 spectrum analysis Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 6
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 240000008042 Zea mays Species 0.000 description 3
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 3
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 3
- 235000005822 corn Nutrition 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 230000003698 anagen phase Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000009342 intercropping Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012271 agricultural production Methods 0.000 description 1
- 235000013339 cereals Nutrition 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Astronomy & Astrophysics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the application provides a method and a system for finely classifying and extracting crops based on satellite remote sensing images, which are combined with a multi-time-sequence remote sensing image crop classification and extraction mode of spectrum, vegetation index and texture characteristics, and the scientificity and reliability of remote sensing ground object classification are greatly improved and the accuracy of crop classification and extraction is improved by comprehensively analyzing the characteristic differences of time and space of ground object types presented on the remote sensing images; a new spectrum index is introduced through spectrum analysis, and secondary classification extraction aiming at target crops is carried out on the basis of obtaining the classification result of the ground objects, so that the classification extraction accuracy of cotton and soybean in the region is improved; by adopting the 2 method result superposition analysis modes, comprehensive algorithm optimization is carried out on the basis of the support vector machine classification result with higher recognition accuracy, so that the classification result can not only keep the accuracy of machine learning and optimizing algorithm classification, but also avoid the limitation of a single algorithm, and meanwhile, the result has more practical significance.
Description
Technical Field
The embodiment of the application relates to the technical field of fine classification of crops, in particular to a fine classification extraction method and system of crops based on satellite remote sensing images.
Background
With the continuous development of technologies such as big data, remote sensing, space information and the Internet of things, the remote sensing images shot by the optical remote sensing satellites are utilized, and the corresponding matching between the images and actual ground objects can be obtained by selecting and analyzing various spectral information and space information features of the ground objects from the images, so that the ground object classification of the remote sensing images is realized. The remote sensing technology can provide various information of the ecological environment of crops and the growth of the crops objectively, accurately and timely, and is an important source for obtaining field data in accurate agriculture. The fine classification of crops has important significance for crop growth monitoring, yield estimation, disaster assessment, national grain safety guarantee and the like, and is an important basis for reasonably distributing resources and accurately fertilizing in the agricultural production process.
The area of the domestic land is wider, the crop planting types are complex and various, and the phenomenon of simultaneous planting of multiple crops of the same species and multiple batches of broken patches and intensive intercropping of crops exists in a certain range of administrative division of the city and county level. Therefore, the difficulty of fine classification of crops is that when the satellite remote sensing image is used for extracting the distribution of the crop areas, the phenomenon of identical foreign matters and foreign matters of the crops are generated due to the fact that the crops show spectrum characteristics with extremely high similarity, so that the crops are extracted to be wrong-separated and missed-separated. At present, by utilizing the traditional remote sensing ground object interpretation technology and combining multiple time sequence satellite images with crop object characteristics, crop differentiation to a certain degree can be realized; crop classification by using unmanned aerial vehicles and hyperspectral remote sensing images and combining deep learning and related network algorithms is also gradually one of the main stream methods of application; meanwhile, various satellite images are subjected to data fusion, and algorithm classification is performed after an image data source with high spatial resolution and multispectral information is obtained, so that the method is one of realization schemes for improving crop identification and classification accuracy. However, from the perspective of data source acquisition, multi-timing satellite images generally correspond to lower spatial resolution, high resolution images lack sufficient time period numbers to reflect the growth phase characteristics of specific crops, and crop classification accuracy is limited; only by means of single climatic features of various crops or single classifiers in machine learning and deep learning classification algorithms, errors caused by human factors are easily introduced in the process of extraction and algorithm parameter setting, and the actual growth condition and the lost classification precision of the crops are ignored while the high-timeliness classification result is obtained.
Disclosure of Invention
The embodiment of the application provides a fine classification extraction method and system for crops based on satellite remote sensing images, which effectively improve the remote sensing classification precision of the crops and better serve the fine classification of the crops by integrating spectrum and texture features with low correlation and high diversity.
In a first aspect, an embodiment of the present application provides a method for finely classifying and extracting crops based on satellite remote sensing images, including:
step S1, acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed, and extracting multi-dimensional features related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional features comprise spectral features, vegetation features and texture features;
s2, inputting the multidimensional features into a pre-trained support vector machine classification model, and determining a classification extraction initial result of the feature class corresponding to the multi-time-series remote sensing image based on the support vector machine classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
and S3, extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
Preferably, the step S1 specifically includes:
step S11, determining multi-time-sequence remote sensing images of the target crops in a period with obvious growth characteristic differences based on the growth period and the growth characteristics of the target crops;
step S12, preprocessing the multi-time sequence remote sensing image, wherein the preprocessing comprises cloud removal, orthographic correction, radiometric calibration, atmospheric correction, image mosaic and clipping and wave band combination;
step S13, superposing and displaying field investigation field crop samples on the preprocessed multi-time sequence images, and establishing a sample spectrum curve; extracting spectral features, vegetation indexes and texture features of the multi-time sequence images; the spectrum characteristics comprise a red wave band, a green wave band, a blue wave band, a near infrared wave band and a short wave infrared wave band of the multi-time sequence remote sensing image; the vegetation characteristics comprise a normalized vegetation index, a ratio vegetation index and an enhanced vegetation index; the texture features include median, covariance, and entropy.
Preferably, the step S2 specifically includes:
screening out one characteristic from the spectral characteristic, the vegetation index and the texture characteristic, calculating the related wave bands of the multi-time sequence image to obtain a characteristic image with the corresponding characteristic, and combining the wave bands of the characteristic images to form an input image;
and inputting the input image into a pre-trained support vector machine classification model, and determining the classification extraction initial result of the land utilization corresponding to the multi-time-sequence remote sensing image based on the support vector machine classification model.
Preferably, the method further comprises:
selecting 80% of the number of various samples from field crop samples for field investigation, and carrying out training of a support vector machine algorithm classifier by combining an input image to obtain a support vector machine classification model for identifying each field crop category; the support vector machine classification model is used for identifying forest lands, water bodies, construction lands, rice, cotton and soybeans in the multi-time sequence remote sensing image; the support vector machine classification model is applied to the test of the whole input image.
Preferably, in the step S2, a feature is selected from each of the spectral feature, the vegetation index and the texture feature, and the relevant wave bands of the multi-time-series image are calculated to obtain a feature image of the corresponding feature, which specifically includes:
extracting an 8-month green wave band in the spectral characteristics, a 9-month enhanced vegetation index in the vegetation characteristics and an 8-month entropy in the texture characteristics as screening characteristics;
the enhanced vegetation index is:
EVI＝2.5*(NIR-R)/(NIR+6*R-7.5*B+1)
in the above formula, NIR, R and B are respectively near infrared band, red band and blue band of the multi-time sequence remote sensing image;
performing 45-degree directional component determination entropy on a first wave band extracted after main component analysis on the multi-time-sequence remote sensing image;
p i the pixel with the gray value of i in the multi-time sequence remote sensing image occupies a proportion.
Preferably, the step S3 specifically includes:
step S31, taking the pattern spots corresponding to the cotton and the soybean in the initial classification and extraction result as a mask, and extracting partial images corresponding to the pattern spots corresponding to the cotton and the soybean in the multi-time-sequence remote sensing image;
s32, extracting a red-edge wave band of the partial image, and determining the spectral reflectivity on the red-edge wave band; selecting two red edge wave bands with the largest difference of spectral reflectivities as a first red edge wave band VRE 1 And a second red edge band VRE 2 Determining a first red edge band VRE 1 And a second red edge band VRE 2 Is a spectrum index of (2):
CSSDI＝(VRE 2 -VRE 1 )/(VRE 2 +VRE 1 )
and step S33, if the spectral index is judged to be smaller than the preset threshold, judging that the pixels corresponding to part of the images are classified as soybeans, otherwise, judging that the pixels corresponding to part of the images are classified as cotton.
Preferably, after step S3, the method further comprises:
s4, carrying out overlay analysis on the fine-classification extraction result of the cotton and the soybean obtained in the step S3 and the initial classification extraction result of the crops to obtain updated fine classification extraction result of the target crops containing six ground object categories and distinguishing the cotton and the soybean;
and S5, carrying out maximum value analysis and cluster analysis on the fine classification extraction result of the target crop to obtain a final classification extraction optimization result of the crop.
In a second aspect, an embodiment of the present application provides a system for finely classifying and extracting crops based on satellite remote sensing images, including:
the characteristic extraction module is used for acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed and extracting multi-dimensional characteristics related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional characteristics comprise spectral characteristics, vegetation characteristics and texture characteristics;
the initial extraction module is used for inputting the multidimensional features into a pre-trained support vector machine classification model, and determining a classification extraction initial result of the feature class corresponding to the multi-time-sequence remote sensing image based on the support vector machine classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
the fine extraction module is used for extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
In a third aspect, an embodiment of the present application 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 the steps of the method for finely classifying and extracting crops based on satellite remote sensing images according to the embodiment of the first aspect of the present application when the processor executes the program.
In a fourth aspect, an embodiment of the present application provides a non-transitory computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the method for finely classifying and extracting crops based on satellite remote sensing images according to the embodiment of the first aspect of the present application.
According to the crop fine classification extraction method and system based on the satellite remote sensing image, provided by the embodiment of the application, the multi-time-sequence remote sensing image crop classification extraction mode of spectrum, vegetation index and texture features is combined, and the scientificity and reliability of remote sensing ground object classification are greatly improved and the accuracy of crop classification extraction is improved by comprehensively analyzing the characteristic differences of time and space of ground object types presented on the remote sensing image; a new spectrum index is introduced through spectrum analysis, and secondary classification extraction aiming at target crops is carried out on the basis of obtaining the classification result of the ground objects, so that the classification extraction accuracy of cotton and soybean in the region is improved; by adopting the 2 method result superposition analysis modes, comprehensive algorithm optimization is carried out on the basis of the support vector machine classification result with higher recognition accuracy, so that the classification result can not only keep the accuracy of machine learning and optimizing algorithm classification, but also avoid the limitation of a single algorithm, and meanwhile, the result has more practical significance.
Drawings
In order to more clearly illustrate the embodiments of the present application 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 present application, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for finely classifying and extracting crops based on satellite remote sensing images, which is provided by the embodiment of the application;
FIG. 2 is a flow chart of the operation of an embodiment of the present application;
fig. 3 is a diagram of cotton and soybean synchronous field planting and remote sensing images provided by an embodiment of the application;
fig. 4 is an initial classification of various extracted feature images of crops according to an embodiment of the present application;
FIG. 5 is a graph of initial results of crop classification extraction provided by an embodiment of the present application;
FIG. 6 is a graph comparing the initial results of classification extraction of cotton and soybean with the classification results of decision trees provided by the embodiment of the application;
FIG. 7 is a block flow diagram of a method for finely classifying and extracting crops based on satellite remote sensing images according to another embodiment of the present application;
fig. 8 is a schematic diagram of an entity structure of an electronic device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The method is one of realization schemes for improving the recognition and classification precision of crops by carrying out data fusion on various satellite images, obtaining image data sources with high spatial resolution and multispectral information and then carrying out algorithm classification. However, from the perspective of data source acquisition, multi-timing satellite images generally correspond to lower spatial resolution, high resolution images lack sufficient time period numbers to reflect the growth phase characteristics of specific crops, and crop classification accuracy is limited; only by means of single climatic features of various crops or single classifiers in machine learning and deep learning classification algorithms, errors caused by human factors are easily introduced in the process of extraction and algorithm parameter setting, and the actual growth condition and the lost classification precision of the crops are ignored while the high-timeliness classification result is obtained.
Therefore, the embodiment of the application provides a method and a system for finely classifying and extracting crops based on satellite remote sensing images, which are used for combining the actual growth characteristics and rules of the crops in areas according to local conditions by combining the use of multi-time sequence and multi-source satellite remote sensing data, fully utilizing higher spatial resolution and rich spectral information, synthesizing various classification models to obtain the crop classification results under the condition of avoiding redundancy of data and algorithms, weakening the phenomena of 'same substance and different spectrum' and 'same spectrum foreign matter' of the crops on the remote sensing images, and effectively improving the remote sensing classification precision of the crops and better serving the fine classification of the crops by integrating the spectrum and texture characteristics with low correlation and high difference. The following description and description will be made with reference to various embodiments.
Fig. 1 and fig. 2 provide a method for finely classifying and extracting crops based on satellite remote sensing images according to an embodiment of the application, which includes:
step S1, acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed, and extracting multi-dimensional features related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional features comprise spectral features, vegetation features and texture features;
step S11, determining multi-time-sequence remote sensing images of the target crops in a period with obvious growth characteristic differences based on the growth period and the growth characteristics of the target crops;
step S12, preprocessing the multi-time sequence remote sensing image, wherein the preprocessing comprises cloud removal, orthographic correction, radiometric calibration, atmospheric correction, image mosaic and clipping and wave band combination;
step S13, superposing and displaying field investigation field crop samples on the preprocessed multi-time sequence images, and establishing a sample spectrum curve; extracting spectral features, vegetation indexes and texture features of the multi-time sequence images; the spectrum characteristics comprise a red wave band, a green wave band, a blue wave band, a near infrared wave band and a short wave infrared wave band of the multi-time sequence remote sensing image; the vegetation characteristics comprise a normalized vegetation index, a ratio vegetation index and an enhanced vegetation index; the texture features include median, covariance, and entropy.
In the embodiment, firstly, 3 uncorrelated characteristics of spectrum characteristics, vegetation indexes and texture characteristics are combined based on a multi-time-sequence remote sensing image of a Sentinel-2 satellite; after main planted crops and related growing periods in the application area are confirmed through investigation, carrying out related image preprocessing operations such as cloud removal, orthographic correction, radiometric calibration, atmospheric correction, image mosaic, cutting, wave band combination and the like on Sentinel-2 multi-time remote sensing images in the target crop growth characteristic difference obvious period; superposing and displaying field investigation field crop samples on the preprocessed Sentinel-2 satellite image, establishing a sample spectrum curve, screening out one characteristic from the spectrum characteristic, the vegetation index and the texture characteristic through differential observation, calculating relevant wave bands of the image, combining the wave bands of the result to form a single input image, adding 80% of field investigation field crop samples, respectively performing training and testing of a classifier by adopting a support vector machine classification algorithm, and dividing land utilization into six types of woodland, water body, construction land, rice, cotton and soybean to obtain an initial result of crop classification extraction; in the embodiment, aiming at the problem of low reliability of crop classification by means of single feature extraction, the multi-time-sequence remote sensing image crop classification extraction mode of spectrum, vegetation index and texture features is combined, and the scientificity and reliability of remote sensing ground object classification are greatly improved by comprehensively analyzing the feature differences of time and space of ground object types on the remote sensing image, so that the accuracy of crop classification extraction is improved.
Selecting 80% of the number of various samples from field crop samples for field investigation, and carrying out training of a support vector machine algorithm classifier by combining an input image to obtain a support vector machine classification model for identifying each field crop category; the support vector machine classification model is used for identifying forest lands, water bodies, construction lands, rice, cotton and soybeans in the multi-time sequence remote sensing image; the support vector machine classification model is applied to the test of the whole input image.
Through field investigation in certain market, the main dominant crops in the market comprise cotton, rice, corn, soybean and the like. In the late 8 th ten days, various crops are in a vigorous development period, the satellite images are obviously different from other types of ground objects, and the crops are green to different degrees; the crops with advantages in the late 9 months generate larger growth difference, for example, cotton is harvested in batches after the cotton is ripened in the late 9 months and is continued for 10 months, corn is gradually harvested until the growth is stopped and the stem is pulled out, the rice still presents dark green similar to that of paddy fields, the soybean is similar to the cotton, and intercropping phenomenon exists mostly, as shown in figure 3. Therefore, in the remote sensing classification extraction process of the crops in the market, the rice and the corn are easy to distinguish and extract from cotton, soybean and other non-crop ground object categories due to obvious time sequence image characteristics, and the cotton and the soybean are difficult to distinguish due to the high similarity of growth cycle and climatic period. Therefore, the image preprocessing operation of cloud removal, orthographic correction, radiometric calibration, atmospheric correction, image mosaic, clipping, band combination and other relevant images is carried out on the images by taking the 8-month late ten-day and 9-month late two-period multi-time sequence Sentinel-2 satellite images as the original data.
S2, inputting the multidimensional features into a pre-trained support vector machine classification model, and determining a classification extraction initial result of the feature class corresponding to the multi-time-series remote sensing image based on the support vector machine classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
screening out one characteristic from the spectral characteristic, the vegetation index and the texture characteristic, calculating the related wave bands of the multi-time sequence image to obtain a characteristic image with the corresponding characteristic, and combining the wave bands of the characteristic images to form an input image;
and (3) superposing and displaying field investigation field crop samples on the two-stage Sentinel-2 satellite images (multi-time sequence remote sensing images) after pretreatment, respectively establishing sample spectrum curves of various indexes under the characteristics of spectrums (red, green, blue, near infrared, short wave infrared bands and the like of the images), vegetation indexes (normalized vegetation index NDVI (Normalized difference vegetation index), specific vegetation index RVI (Ratio vegetation index), enhanced vegetation index EVI (Enhanced vegetation index) and the like) and textures (median, covariance, entropy and the like), and carrying out selection of various single indexes with low correlation through differential observation and index calculation principle analysis. Finally, the characteristic spectrum, the 8-month green wave band, the vegetation index, the 9-month enhanced vegetation index, the texture and the 8-month entropy are selected, and each characteristic image is shown in the figure 4. The green wave band can be directly obtained through wave band extraction; the calculation formula of the enhanced vegetation index is as follows:
EVI＝2.5*(NIR-R)/(NIR+6*R-7.5*B+1)
in the above formula, NIR, R and B are respectively near infrared band, red band and blue band of the multi-time sequence remote sensing image;
performing 45-degree directional component determination entropy on a first wave band extracted after main component analysis on the multi-time-sequence remote sensing image;
p i the pixel with the gray value of i in the multi-time sequence remote sensing image occupies a proportion.
And inputting the input image into a pre-trained support vector machine classification model, and determining the classification extraction initial result of the land utilization corresponding to the multi-time-sequence remote sensing image based on the support vector machine classification model. And carrying out wave band combination on each characteristic image to form a single input image result. In field investigation field crop samples, 80% of the number of various samples is selected, and training of a support vector machine algorithm classifier is carried out by combining an input image, so that a sample classification model of each field crop category is obtained. The model is applied to the test of the whole input image, and the initial result of six types of land classification extraction of woodland, water body, construction land, rice, cotton and soybean based on the sample characteristic model is obtained, and the result is shown in figure 5. In the embodiment, the 3 characteristics of spectrum, vegetation index and texture are combined to carry out fine classification of crops supporting a vector machine algorithm, the classification result precision is higher, and the model usability is stronger.
And S3, extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
Through comparing the initial results with government statistical data, the cotton and the soybean have more false and missed phenomena. Combining red Bian Boduan of the preprocessed Sentinel-2 satellite image, carrying out different new spectral feature extraction on images corresponding to initial classification results of cotton and soybean, and realizing cotton and soybean image extraction based on red-edge wave bands through a decision tree classification algorithm with a set threshold value; aiming at the problems that cotton and soybean crops are similar in growth period and climatic period and difficult to distinguish and extract on remote sensing images through algorithm classification, the embodiment of the application provides a distinguishing and extracting method based on satellite image red-edge band characteristics, a new spectrum index is introduced through spectrum analysis, secondary classification and extraction on target crops are carried out on the basis of obtaining ground object classification results, and the classification and extraction accuracy of regional cotton and soybean is improved.
And carrying out fine differential extraction of cotton and soybean on the basis of the initial result of the classified extraction. At the beginning of 10 months at the end of 9 months, soybeans in certain markets are ripe and harvested, cotton also enters a boll picking period, and certain differences exist in corresponding images of crops. Since the satellite image red band value is closely related to vegetation growth state, by counting spectral reflectances of cotton and soybean samples on the 1 red band and the 3 red bands of the 9-month Sentinel-2 satellite image, it is found that the spectral reflectances of the two crops have obvious differences in the 1 red band (the first red band) and the 2 red band (the second red band).
S31, using cotton and soybean pattern spots in the initial result of classification extraction as a mask, and extracting partial images corresponding to the initial cotton and soybean pattern spots in the original 9-month Sentinel-2 satellite image;
and S32, extracting red-edge band related spectral features of cotton and soybean images to amplify the spectral difference of the two crops for subsequent feature classification. Determining a spectral reflectance over the red band; selecting two red edge wave bands with the largest difference of spectral reflectivities as a first red edge wave band VRE 1 And a second red edge band VRE 2 Determining a first red edge band VRE 1 And a second red edge band VRE 2 Is a spectrum index of (2):
CSSDI＝(VRE 2 -VRE 1 )/(VRE 2 +VRE 1 )
and step S33, if the spectral index is judged to be smaller than the preset threshold, judging that the pixels corresponding to part of the images are classified as soybeans, otherwise, judging that the pixels corresponding to part of the images are classified as cotton.
And (3) counting CSSDI values of cotton and soybean samples on the single-band result image obtained in the step (3 b), and selecting 0.299 as a segmentation threshold value to classify ground features of a decision tree algorithm. The decision tree structure is as follows: when CSSDI > =0.299, the pixel is classified as cotton; when CSSDI <0.299, the pixels are classified as soybean. The cotton and soybean classification extracts initial results and decision tree classification results are shown in figure 6, for example. Based on the red-edge band characteristics of satellite images, a new spectrum index (CSSDI) is introduced through spectrum analysis, and the cotton and soybean can be distinguished and extracted on the basis of the result with better classification precision, so that the high-precision fine classification of crops with regional dominant crops as targets is realized; the method combines a support vector machine algorithm and a decision tree algorithm to carry out fine classification of crops, avoids the problems of low classification reliability and insufficient classification precision of a single algorithm which purely depends on machine learning, and improves the classification precision to be applied to agriculture.
Fig. 7 is a flowchart of a method for finely classifying and extracting crops based on satellite remote sensing images according to another embodiment of the present application, referring to fig. 7, after step S3, the method further includes:
s4, carrying out overlay analysis on the fine-classification extraction result of the cotton and the soybean obtained in the step S3 and the initial classification extraction result of the crops to obtain updated fine classification extraction result of the target crops containing six ground object categories and distinguishing the cotton and the soybean;
and S5, carrying out remote sensing classification post-processing operations such as maximum value analysis, cluster analysis, manual modification based on manual visual interpretation and the like on the fine classification extraction result of the target crop to obtain a final crop classification extraction optimization result. And then combining 20% of field investigation field crop samples, and performing confusion matrix classification precision evaluation based on real ground object reference, wherein the overall precision, kappa coefficient, production precision and user precision are used as indexes. The result shows that the overall classification accuracy of nearly 90% can be realized by classifying the ground features of the multi-time sequence images by combining the spectrum, the vegetation index and the texture features and using a support vector machine classification algorithm, but the extraction accuracy of cotton and soybean crops is lower than 70%; after the CSSDI index is increased and a decision tree classification algorithm is combined, the overall accuracy of the ground feature classification is improved to more than 95%, and the accuracy is improved by 5.39% compared with that of a single algorithm; the extraction precision index values of cotton and soybean are substantially improved to 85% or above, wherein the production precision and the user precision are respectively improved by 20.6%, 25.73%, 34.43% and 25.81%. The detailed classification accuracy is compared with the following table 1.
TABLE 1 precision comparison of crop Fine Classification
In an embodiment, the embodiment of the application further provides a system for finely classifying and extracting crops based on satellite remote sensing images, and the method for finely classifying and extracting crops based on satellite remote sensing images in the above embodiments comprises the following steps:
the characteristic extraction module is used for acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed and extracting multi-dimensional characteristics related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional characteristics comprise spectral characteristics, vegetation characteristics and texture characteristics;
the primary extraction module is used for inputting the multidimensional features into a pre-trained primary classification model, and determining a classification extraction initial result of the ground object category corresponding to the multi-time-sequence remote sensing image based on the primary classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
the fine extraction module is used for extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
In one embodiment, the present application further provides an electronic device, as shown in fig. 8, which 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. The processor 810 may invoke logic instructions in the memory 830 to perform the steps of the satellite telemetry image based crop fine classification extraction method as described in the embodiments above. Examples include:
step S1, acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed, and extracting multi-dimensional features related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional features comprise spectral features, vegetation features and texture features;
s2, inputting the multidimensional features into a pre-trained support vector machine classification model, and determining a classification extraction initial result of the feature class corresponding to the multi-time-series remote sensing image based on the support vector machine classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
and S3, extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
In one embodiment, the embodiment of the present application further provides a non-transitory computer readable storage medium, where the computer readable storage medium stores a computer program, where the computer program includes at least one piece of code, where the at least one piece of code is executable by a master control device to control the master control device to implement the steps of the method for finely classifying and extracting crops based on satellite remote sensing images according to the embodiments above. Examples include:
step S1, acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed, and extracting multi-dimensional features related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional features comprise spectral features, vegetation features and texture features;
s2, inputting the multidimensional features into a pre-trained support vector machine classification model, and determining a classification extraction initial result of the feature class corresponding to the multi-time-series remote sensing image based on the support vector machine classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
and S3, extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
In one embodiment, the present application also provides a computer program for implementing the above method embodiment when the computer program is executed by a master control device.
The program may be stored in whole or in part on a storage medium that is packaged with the processor, or in part or in whole on a memory that is not packaged with the processor.
In one embodiment, the present application further provides a processor, where the processor is configured to implement the above method embodiment. The processor may be a chip.
In summary, according to the method and the system for finely classifying and extracting the crops based on the satellite remote sensing image provided by the embodiment of the application, the spectrum, the vegetation index and the texture characteristic are combined in the multi-time-sequence remote sensing image crop classification and extraction mode, and the scientificity and the reliability of remote sensing ground object classification are greatly improved by comprehensively analyzing the characteristic differences of time and space of the ground object types on the remote sensing image, so that the accuracy of crop classification and extraction is improved; a new spectrum index is introduced through spectrum analysis, and secondary classification extraction aiming at target crops is carried out on the basis of obtaining the classification result of the ground objects, so that the classification extraction accuracy of cotton and soybean in the region is improved; by adopting the 2 method result superposition analysis modes, comprehensive algorithm optimization is carried out on the basis of the support vector machine classification result with higher recognition accuracy, so that the classification result can not only keep the accuracy of machine learning and optimizing algorithm classification, but also avoid the limitation of a single algorithm, and meanwhile, the result has more practical significance.
In the foregoing embodiments, the descriptions of the embodiments are focused on, and for those portions of one embodiment that are not described in detail, reference may be made to the related descriptions of other embodiments.
It will be apparent to those skilled in the art that embodiments of the present application may be provided as a method, apparatus, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (10)
1. The fine classification and extraction method for crops based on satellite remote sensing images is characterized by comprising the following steps of:
step S1, acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed, and extracting multi-dimensional features related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional features comprise spectral features, vegetation features and texture features;
s2, inputting the multidimensional features into a pre-trained support vector machine classification model, and determining a classification extraction initial result of the feature class corresponding to the multi-time-series remote sensing image based on the support vector machine classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
and S3, extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
2. The method for finely classifying and extracting crops based on satellite remote sensing images according to claim 1, wherein the step S1 specifically comprises:
step S11, determining multi-time-sequence remote sensing images of the target crops in a period with obvious growth characteristic differences based on the growth period and the growth characteristics of the target crops;
step S12, preprocessing the multi-time sequence remote sensing image, wherein the preprocessing comprises cloud removal, orthographic correction, radiometric calibration, atmospheric correction, image mosaic and clipping and wave band combination;
step S13, superposing and displaying field investigation field crop samples on the preprocessed multi-time sequence images, and establishing a sample spectrum curve; extracting spectral features, vegetation indexes and texture features of the multi-time sequence images; the spectrum characteristics comprise a red wave band, a green wave band, a blue wave band, a near infrared wave band and a short wave infrared wave band of the multi-time sequence remote sensing image; the vegetation characteristics comprise a normalized vegetation index, a ratio vegetation index and an enhanced vegetation index; the texture features include median, covariance, and entropy.
3. The method for finely classifying and extracting crops based on satellite remote sensing images according to claim 2, wherein the step S2 specifically comprises:
screening out one characteristic from the spectral characteristic, the vegetation index and the texture characteristic, calculating the related wave bands of the multi-time sequence image to obtain a characteristic image with the corresponding characteristic, and combining the wave bands of the characteristic images to form an input image;
and inputting the input image into a pre-trained support vector machine classification model, and determining the classification extraction initial result of the land utilization corresponding to the multi-time-sequence remote sensing image based on the support vector machine classification model.
4. The method for finely classifying and extracting crops based on satellite remote sensing images according to claim 2, further comprising:
selecting 80% of the number of various samples from field crop samples for field investigation, and carrying out training of a support vector machine algorithm classifier by combining an input image to obtain a support vector machine classification model for identifying each field crop category; the support vector machine classification model is used for identifying forest lands, water bodies, construction lands, rice, cotton and soybeans in the multi-time sequence remote sensing image; the support vector machine classification model is applied to the test of the whole input image.
5. The method for finely classifying and extracting crops based on satellite remote sensing images according to claim 3, wherein in the step S2, one feature is selected from each of spectral features, vegetation indexes and texture features, and the relevant wave bands of the multi-time-series images are calculated to obtain feature images of the corresponding features, which specifically comprises:
extracting an 8-month green wave band in the spectral characteristics, a 9-month enhanced vegetation index in the vegetation characteristics and an 8-month entropy in the texture characteristics as screening characteristics;
the enhanced vegetation index is:
EVI＝2.5*(NIR-R)/(NIR+6*R-7.5*B+1)
in the above formula, NIR, R and B are respectively near infrared band, red band and blue band of the multi-time sequence remote sensing image;
performing 45-degree directional component determination entropy on a first wave band extracted after main component analysis on the multi-time-sequence remote sensing image;
p i the pixel with the gray value of i in the multi-time sequence remote sensing image occupies a proportion.
6. The method for finely classifying and extracting crops based on satellite remote sensing images according to claim 1, wherein the step S3 specifically comprises:
step S31, taking the pattern spots corresponding to the cotton and the soybean in the initial classification and extraction result as a mask, and extracting partial images corresponding to the pattern spots corresponding to the cotton and the soybean in the multi-time-sequence remote sensing image;
s32, extracting a red-edge wave band of the partial image, and determining the spectral reflectivity on the red-edge wave band; selecting two red edge wave bands with the largest difference of spectral reflectivities as a first red edge wave band VRE 1 And a second red edge band VRE 2 Determining a first red edge band VRE 1 And a second red edge band VRE 2 Is a spectrum index of (2):
CSSDI＝(VRE 2 -VRE 1 )/(VRE 2 +VRE 1 )
and step S33, if the spectral index is judged to be smaller than the preset threshold, judging that the pixels corresponding to part of the images are classified as soybeans, otherwise, judging that the pixels corresponding to part of the images are classified as cotton.
7. The method for finely classifying and extracting crops based on satellite remote sensing images according to claim 2, wherein after step S3, the method further comprises:
s4, carrying out overlay analysis on the fine-classification extraction result of the cotton and the soybean obtained in the step S3 and the initial classification extraction result of the crops to obtain updated fine classification extraction result of the target crops containing six ground object categories and distinguishing the cotton and the soybean;
and S5, carrying out maximum value analysis and cluster analysis on the fine classification extraction result of the target crop to obtain a final classification extraction optimization result of the crop.
8. The utility model provides a fine classification extraction system of crops based on satellite remote sensing image which characterized in that includes:
the characteristic extraction module is used for acquiring a multi-time-sequence remote sensing image of crops in an area to be analyzed and extracting multi-dimensional characteristics related to the types of the crops in the multi-time-sequence remote sensing image, wherein the multi-dimensional characteristics comprise spectral characteristics, vegetation characteristics and texture characteristics;
the initial extraction module is used for inputting the multidimensional features into a pre-trained support vector machine classification model, and determining a classification extraction initial result of the feature class corresponding to the multi-time-sequence remote sensing image based on the support vector machine classification model; the initial result of the classification extraction comprises rice, cotton, soybean and non-crop categories;
the fine extraction module is used for extracting multi-time-sequence remote sensing images corresponding to cotton and soybean in the initial classification and extraction result, extracting red-edge wave bands in the multi-time-sequence remote sensing images corresponding to the cotton and the soybean through a decision tree classification algorithm with a set segmentation threshold, determining a secondary classification and extraction result of the cotton and the soybean based on the red-edge wave bands, and correcting the initial classification and extraction result based on the secondary classification and extraction result to obtain a fine classification and extraction result of crops of the multi-time-sequence remote sensing images.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor performs the steps of the satellite remote sensing image based crop fine classification extraction method of any of claims 1 to 7 when the program is executed.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor, performs the steps of the satellite remote sensing image based crop fine classification extraction method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310847822.2A CN117079152A (en) | 2023-07-11 | 2023-07-11 | Fine crop classification extraction method and system based on satellite remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310847822.2A CN117079152A (en) | 2023-07-11 | 2023-07-11 | Fine crop classification extraction method and system based on satellite remote sensing image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117079152A true CN117079152A (en) | 2023-11-17 |
Family
ID=88701242
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310847822.2A Pending CN117079152A (en) | 2023-07-11 | 2023-07-11 | Fine crop classification extraction method and system based on satellite remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117079152A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117292267A (en) * | 2023-11-27 | 2023-12-26 | 武汉大学 | Method and system for estimating rice aboveground biomass in segments based on weather information |
| CN117789023A (en) * | 2023-12-26 | 2024-03-29 | 江苏省金威遥感数据工程有限公司 | Remote sensing identification system of crop planting structure |
| CN118470441A (en) * | 2024-07-10 | 2024-08-09 | 广东省科学院广州地理研究所 | Multi-time-phase high-resolution remote sensing image-based farmland classification method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020635A (en) * | 2019-04-15 | 2019-07-16 | 中国农业科学院农业资源与农业区划研究所 | Growing area crops sophisticated category method and system based on unmanned plane image and satellite image |
| CN113901966A (en) * | 2021-12-07 | 2022-01-07 | 武汉光谷信息技术股份有限公司 | Crop classification method fusing multi-source geographic information data |
| CN114612794A (en) * | 2022-03-01 | 2022-06-10 | 中国农业大学 | Remote sensing identification method for land covering and planting structure in finely-divided agricultural area |
| WO2023000160A1 (en) * | 2021-07-20 | 2023-01-26 | 海南长光卫星信息技术有限公司 | Hyperspectral remote sensing image semi-supervised classification method, apparatus, and device, and storage medium |
-
2023
- 2023-07-11 CN CN202310847822.2A patent/CN117079152A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020635A (en) * | 2019-04-15 | 2019-07-16 | 中国农业科学院农业资源与农业区划研究所 | Growing area crops sophisticated category method and system based on unmanned plane image and satellite image |
| WO2023000160A1 (en) * | 2021-07-20 | 2023-01-26 | 海南长光卫星信息技术有限公司 | Hyperspectral remote sensing image semi-supervised classification method, apparatus, and device, and storage medium |
| CN113901966A (en) * | 2021-12-07 | 2022-01-07 | 武汉光谷信息技术股份有限公司 | Crop classification method fusing multi-source geographic information data |
| CN114612794A (en) * | 2022-03-01 | 2022-06-10 | 中国农业大学 | Remote sensing identification method for land covering and planting structure in finely-divided agricultural area |
Non-Patent Citations (3)
| Title |
|---|
| YONG HONG, DEREN LI, MI WANG, HAONAN JIANG, LENGKUN LUO, YANPING WU, CHEN LIU, TIANJIN XIE, QING ZHANG AND ZAHID JAHANGIR: "Cotton Cultivated Area Extraction Based on Multi-Feature Combination and CSSDI under Spatial Constraint", REMOTE SENSING, vol. 14, no. 6, 13 March 2022 (2022-03-13), pages 1 - 20 * |
| 史飞飞;雷春苗;肖建设;李甫;石明明;: "基于多源遥感数据的复杂地形区农作物分类", 地理与地理信息科学, vol. 34, no. 05, 30 September 2018 (2018-09-30) * |
| 王云艳; 罗冷坤; 周志刚;: "改进型DeepLab的极化SAR果园分类", 中国图象图形学报, vol. 24, no. 11, 16 November 2019 (2019-11-16) * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117292267A (en) * | 2023-11-27 | 2023-12-26 | 武汉大学 | Method and system for estimating rice aboveground biomass in segments based on weather information |
| CN117292267B (en) * | 2023-11-27 | 2024-02-02 | 武汉大学 | Method and system for estimating rice aboveground biomass in segments based on weather information |
| CN117789023A (en) * | 2023-12-26 | 2024-03-29 | 江苏省金威遥感数据工程有限公司 | Remote sensing identification system of crop planting structure |
| CN117789023B (en) * | 2023-12-26 | 2024-10-18 | 江苏省金威遥感数据工程有限公司 | Remote sensing identification system of crop planting structure |
| CN118470441A (en) * | 2024-07-10 | 2024-08-09 | 广东省科学院广州地理研究所 | Multi-time-phase high-resolution remote sensing image-based farmland classification method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Barbedo et al. | Identifying multiple plant diseases using digital image processing | |
| George et al. | Forest tree species discrimination in western Himalaya using EO-1 Hyperion | |
| Paulus et al. | Surface feature based classification of plant organs from 3D laserscanned point clouds for plant phenotyping | |
| CN117079152A (en) | Fine crop classification extraction method and system based on satellite remote sensing image | |
| Puletti et al. | Unsupervised classification of very high remotely sensed images for grapevine rows detection | |
| Triscowati et al. | Classification of rice-plant growth phase using supervised random forest method based on landsat-8 multitemporal data | |
| Ninomiya | High-throughput field crop phenotyping: current status and challenges | |
| CN113029971B (en) | Crop canopy nitrogen monitoring method and system | |
| CN108710864B (en) | Winter wheat remote sensing extraction method based on multi-dimensional identification and image noise reduction processing | |
| CN109977901A (en) | A kind of corn weeds in field recognition methods based on unmanned aerial vehicle remote sensing | |
| CN113657158B (en) | Google EARTH ENGINE-based large-scale soybean planting area extraction algorithm | |
| CN113252584B (en) | Crop growth detection method and system based on 5G transmission | |
| Khoshroo et al. | Detection of red tomato on plants using image processing techniques. | |
| Netto et al. | Segmentation of RGB images using different vegetation indices and thresholding methods | |
| CN116543316B (en) | Method for identifying turf in paddy field by utilizing multi-time-phase high-resolution satellite image | |
| Chen et al. | A novel Greenness and Water Content Composite Index (GWCCI) for soybean mapping from single remotely sensed multispectral images | |
| Kumar et al. | An identification of crop disease using image segmentation | |
| Panda et al. | Distinguishing blueberry bushes from mixed vegetation land use using high resolution satellite imagery and geospatial techniques | |
| Shu et al. | Assessing maize lodging severity using multitemporal UAV-based digital images | |
| Roth et al. | Automated detection of individual clove trees for yield quantification in northeastern Madagascar based on multi-spectral satellite data | |
| Barreto et al. | Disease incidence and severity of cercospora leaf spot in sugar beet assessed by multispectral unmanned aerial images and machine learning | |
| Mendigoria et al. | Seed architectural phenes prediction and variety classification of dry beans (phaseolus vulgaris) using machine learning algorithms | |
| Rony et al. | BottleNet18: Deep Learning-Based Bottle Gourd Leaf Disease Classification | |
| CN115063678A (en) | Soybean planting area extraction method based on domestic GF-6WFV data | |
| Zhao et al. | A multispectral image based object detection approach in natural scene |
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 |