CN113989657A - Method and device for detecting farmland range change based on invariant information sample screening - Google Patents
Method and device for detecting farmland range change based on invariant information sample screening Download PDFInfo
- Publication number
- CN113989657A CN113989657A CN202111184439.0A CN202111184439A CN113989657A CN 113989657 A CN113989657 A CN 113989657A CN 202111184439 A CN202111184439 A CN 202111184439A CN 113989657 A CN113989657 A CN 113989657A
- Authority
- CN
- China
- Prior art keywords
- land
- image
- screening
- samples
- change detection
- 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.)
- Granted
Links
- 230000008859 change Effects 0.000 title claims abstract description 101
- 238000012216 screening Methods 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000001514 detection method Methods 0.000 claims abstract description 89
- 230000003595 spectral effect Effects 0.000 claims abstract description 29
- 230000011218 segmentation Effects 0.000 claims abstract description 28
- 238000002474 experimental method Methods 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000013145 classification model Methods 0.000 claims abstract description 8
- 238000000605 extraction Methods 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 19
- 239000013598 vector Substances 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 16
- 238000001228 spectrum Methods 0.000 claims description 15
- 238000009826 distribution Methods 0.000 claims description 13
- 238000011156 evaluation Methods 0.000 claims description 12
- 230000014509 gene expression Effects 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 238000007689 inspection Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 2
- 238000003971 tillage Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 10
- 238000013135 deep learning Methods 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 238000010276 construction Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 235000002566 Capsicum Nutrition 0.000 description 1
- 239000006002 Pepper Substances 0.000 description 1
- 235000016761 Piper aduncum Nutrition 0.000 description 1
- 235000017804 Piper guineense Nutrition 0.000 description 1
- 244000203593 Piper nigrum Species 0.000 description 1
- 235000008184 Piper nigrum Nutrition 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000011895 specific detection Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010897 surface acoustic wave method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a farmland range change detection method based on invariant information sample screening, which comprises the following steps: extracting classification information: acquiring raster images of various places and determining a region to be detected; extracting spectral and textural features: obtaining an image object, and extracting spectral features and textural features; screening samples: the superpixels obtained by segmentation are used as initial samples, and experimental samples are screened based on invariant information; selecting the optimal characteristics: carrying out classification experiments on different land samples to determine image characteristic differences between cultivated land and other land samples; detecting a change area: training an SVM classification model by using an experimental sample, determining a kernel function parameter of the model, selecting an optimal image characteristic, classifying superpixel segmentation results of a farmland range region to be detected, extracting a changed superpixel object, and detecting a farmland change region. The method realizes automatic sample screening by using the invariant information in the existing data, and effectively detects the farmland change information of the high-resolution remote sensing image.
Description
Technical Field
The invention relates to a method and a device for detecting farmland range change based on invariant information sample screening, and belongs to the technical field of remote sensing image processing.
Background
The remote sensing image change detection is to determine and analyze the change of the ground features in the area, including the change of the position, range, property and state of the ground features, by using the multi-source remote sensing data and the related geographic space data of different time phases in the same area, combining the corresponding ground feature characteristics and the remote sensing imaging principle and adopting an image and graphic processing theory and a mathematical model method. Scholars at home and abroad make a great deal of research on the remote sensing image change detection method and obtain a lot of achievements.
According to the unit division of change detection, the high-resolution remote sensing image change detection method can be divided into two types, namely a pixel-based method and an object-oriented method. The former takes picture elements as a unit, ignores the spatial correlation among the picture elements in the image and is easy to generate salt and pepper noise; the latter takes an image object as a unit, can fully utilize the spatial relationship in the image, and has been widely applied to the change detection of high-resolution remote sensing images.
With the development of artificial intelligence, the application of the method in the field of remote sensing change detection is more and more extensive, and the change detection research based on machine learning/deep learning is vigorous. Machine learning and deep learning can directly obtain change detection results from multi-temporal remote sensing images, and the problems that the precision of the results of the traditional change detection method depends on a difference graph greatly and the precision of the detection results is unstable are solved to a certain extent. However, the change detection based on the machine learning/deep learning requires a large amount of sample data for training and a complete sample library is constructed, and the construction of the sample library is a time-consuming and labor-consuming work, and is a problem to be solved urgently in the machine learning/deep learning change detection.
In addition, in recent years, the 'non-agricultural' behavior of the cultivated land such as building houses in a disorderly manner, planting trees and forestation, digging lakes and landscaping is frequently prohibited and even has a more advanced trend, so that serious threats are caused to the cultivated land protection work of China. The state puts forward greater demands and higher requirements on the supervision of natural resources, particularly cultivated land resources, and needs to utilize remote sensing technology and means to carry out accurate dynamic monitoring on cultivated land to accurately grasp the utilization and change conditions of the cultivated land. However, the existing research for detecting the change of the farmland by using the remote sensing image mainly has the problems of low efficiency, insufficient precision and the like, and most of researches do not consider how to mine and use the invariant information in the existing data result to assist in detecting the change of the farmland. Therefore, the research on the farmland change detection aspect has a large blank, and is also a great vacancy of the national farmland resource supervision and monitoring requirements.
Disclosure of Invention
The invention aims to provide a method and a device for detecting farmland range change based on invariant information sample screening, which can realize automatic sample screening by utilizing invariant information in the existing data, can effectively detect farmland change information of a high-resolution remote sensing image, and can solve the problem of difficult construction of a sample library in the process of change detection by utilizing machine/deep learning at present.
Based on the same invention concept, the invention has four independent technical schemes:
in a first aspect, the arable land range change detection method based on invariant information sample screening provided by the embodiment of the invention comprises the following steps:
extracting classification information: analyzing and processing the homeland by utilizing survey data, acquiring grid images corresponding to all land types, and determining the acquired farmland range images as the to-be-detected area;
extracting spectral and textural features: performing super-pixel segmentation on the grid image of the region to be detected to obtain an image object, and extracting the spectral feature and the texture feature of the image object;
screening samples: randomly selecting a certain number of grid images from grid images corresponding to various places, carrying out superpixel segmentation, taking the obtained superpixels as initial samples of the various places, and screening experimental samples;
selecting the optimal characteristics: selecting different image characteristics, performing classification experiments on different land samples obtained through automatic screening, and determining image characteristic differences between cultivated land and other land types;
detecting a change area: training an SVM classification model by using an experimental sample, and adaptively determining kernel function parameters of the model by adopting a grid search method; and selecting the optimal image characteristics according to different farmland change types, classifying the superpixel segmentation results of the to-be-detected region by using a trained SVM model, extracting changed superpixel objects, and detecting the farmland change region.
As a possible implementation manner of this embodiment, the extracting the classification information includes:
obtaining a remote sensing image utilized by the homeland;
processing the remote sensing image by using the vector boundary and the category attribute of the image spot to obtain the vector range of each land category;
and performing mask extraction on the obtained remote sensing images of the vector ranges of all the land types to obtain grid images corresponding to all the land types, and determining the range of the area to be detected according to the obtained cultivated land range images.
As a possible implementation manner of this embodiment, the extracting the spectral and textural features includes:
extracting the spectral average value I of the image object in R, G, B three bandsR、IGAnd IBAs a spectral feature value;
calculating a texture characteristic value of the image object by utilizing the gray level co-occurrence matrix;
the texture feature values include angular second moment, contrast, entropy and homogeneity.
As a possible implementation manner of this embodiment, the spectrum mean value IR、IGAnd IBThe expression of (a) is:
wherein n is the number of pixels in each image object, Ii R、Ii G、Ii BRespectively is the spectral value of the ith pixel in the super pixel in R, G, B three bands;
the expression of the texture feature value is as follows:
wherein ASM is angular second moment, CON is contrast, ENT is entropy, HOM is homogeneity, L represents the number of image gray levels, p (i, j) is an element of a gray level co-occurrence matrix, and i, j are row elements and column elements of the matrix elements, respectively.
As a possible implementation manner of this embodiment, the screening the sample includes:
randomly selecting a certain number of pattern spots from the grid images corresponding to each land class, obtaining the grid images of the pattern spots by using a mask extraction method, performing superpixel segmentation on the grid images of the pattern spots of each land class, and taking the obtained superpixels as initial samples of each land class;
calculating the homogeneity characteristic value of each type of initial sample, drawing a box line graph of the distribution of the homogeneity characteristic value, based on the characteristic that the homogeneity characteristic distribution of the type of samples in an unchanged area in an image is concentrated, eliminating abnormal samples with the homogeneity characteristic value larger than an upper limit and smaller than a lower limit by using an abnormal analysis method, and selecting a required number of experimental samples from the rest samples.
As a possible implementation manner of this embodiment, the expressions of the upper bound UB and the lower bound LB of the box line graph are respectively:
UB＝Qu+1.5IQR
LB＝Ql-1.5IQR
where Qu is the upper quartile of the boxplot distribution, Ql is the lower quartile, and IQR is the difference between the upper quartile Qu and the lower quartile Ql.
As a possible implementation manner of this embodiment, the kernel function parameters of the model include C and γ, where C is a penalty coefficient, i.e., a tolerance to errors; gamma is a parameter of the RBF function itself after the function is selected as the kernel function.
As a possible implementation manner of this embodiment, the method further includes the following steps:
evaluating the precision of the detection result; and comparing and analyzing the change detection result with a true value image visually interpreted, selecting 3 indexes of a virtual inspection rate, a missed inspection rate and a correct rate to evaluate the precision, and evaluating the accuracy of the change detection result.
As a possible implementation manner of this embodiment, the expressions of the false detection rate, the missed detection rate, and the correct rate are:
wherein FP is the number of pixels which are actually unchanged and detected as changed, FC is the total number of pixels which are detected as changed, FN is the number of pixels which are actually changed and detected as unchanged, PC is the total number of pixels which are actually changed, TP is the number of pixels which are actually changed and detected as changed, TN is the number of pixels which are actually unchanged and detected as unchanged, and N is the total number of all pixels in the detection area.
In a second aspect, an arable land range change detection device based on invariant information sample screening provided by an embodiment of the present invention includes:
the classification information extraction module is used for analyzing and processing the homeland utilization survey data, acquiring grid images corresponding to all land types, and determining the acquired farmland range images as the to-be-detected area;
the spectrum and texture feature extraction module is used for performing super-pixel segmentation on the grid image of the region to be detected to obtain an image object and extracting the spectrum feature and the texture feature of the image object;
the sample screening module is used for randomly selecting a certain number of grid images from the grid images corresponding to all the places, carrying out superpixel segmentation, taking the obtained superpixels as initial samples of all the places, and screening experimental samples;
the optimal feature selection module is used for selecting different image features, performing classification experiments on different land samples obtained through automatic screening, and determining image feature differences between cultivated land and other land types;
the change region detection module is used for training the SVM classification model by using an experimental sample and adaptively determining kernel function parameters of the model by adopting a grid search method; and selecting the optimal image characteristics according to different farmland change types, classifying the superpixel segmentation results of the to-be-detected region by using a trained SVM model, extracting changed superpixel objects, and detecting the farmland change region.
As a possible implementation manner of this embodiment, the apparatus further includes:
and the detection result evaluation module is used for comparing and analyzing the change detection result with the visually-interpreted true value image, selecting 3 indexes of the virtual inspection rate, the omission factor and the accuracy rate to perform precision evaluation, and evaluating the accuracy of the change detection result.
In a third aspect, the arable land range change detection device based on invariant information sample screening provided by the embodiment of the invention comprises a processor and a memory, wherein the memory stores computer program instructions;
the processor reads and executes the computer program instructions to implement the arable land range change detection method based on invariant information sample screening as described in any of the above.
In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, where the computer program is executed by a processor to perform the steps of any of the above-mentioned methods for detecting farmland range change based on invariant information sample screening.
The invention has the following beneficial effects:
the method can realize automatic sample screening by using the invariant information in the existing data, can effectively detect the farmland change information of the high-resolution remote sensing image, and solves the problem of difficult sample library construction in the change detection process by using machine/deep learning at present.
The object-oriented detection method for the optimal image characteristics is selected according to different change types, and the change detection precision is effectively improved.
Drawings
FIG. 1 is a flow chart illustrating a cultivated land extent change detection method based on invariant information sample screening according to an exemplary embodiment;
FIG. 2 is a block diagram illustrating a cultivated land area change detection apparatus based on invariant information sample screening according to an exemplary embodiment;
FIG. 3 is a schematic view of the farmland range change detection process based on invariant information sample screening by using the farmland range change detection device based on invariant information sample screening according to the present invention;
FIG. 4 is a schematic diagram illustrating a remote sensing image processing according to an exemplary embodiment.
Detailed Description
The present invention is described in detail with reference to the embodiments shown in the drawings, but it should be understood that these embodiments are not intended to limit the present invention, and those skilled in the art should understand that functional, methodological, or structural equivalents or substitutions made by these embodiments are within the scope of the present invention.
The land types comprise cultivated land, garden land, forest land, grassland, business land, industrial and mining storage land, residential land, public management and public service land, special land, transportation land, water and water conservancy facilities land and other land.
The first embodiment is as follows:
as shown in fig. 1, a method for detecting farmland range change based on invariant information sample screening provided by an embodiment of the present invention includes the following steps:
step 1, extracting classification information: analyzing and processing the homeland by utilizing survey data, acquiring the grid images corresponding to all land types, and determining the acquired farmland range images as the to-be-detected area.
Obtaining a remote sensing image of the national utilization from the national utilization survey result of the third time; processing the remote sensing image by using the vector boundary and the category attribute of the image spot to obtain the vector range of each land category; and performing mask extraction on the obtained remote sensing images of the vector ranges of all the land types to obtain grid images corresponding to all the land types, and determining the range of the area to be detected according to the obtained cultivated land range images. The step rapidly positions the region to be detected in the arable land range to be detected, and simultaneously facilitates subsequent selection of samples in various land vector ranges.
Extracting the spectral average value I of the image object in R, G, B three bandsR、IGAnd IBAs a spectral feature value; the spectral mean value IR、IGAnd IBThe expression of (a) is:
wherein n is the number of pixels in each image object, Ii R、Ii G、Ii BRespectively is the spectral value of the ith pixel in the super pixel in R, G, B three bands;
calculating a texture characteristic value of the image object by utilizing the gray level co-occurrence matrix; the texture characteristic values comprise angular second moment, contrast, entropy and homogeneity, and the expressions are as follows:
wherein ASM is angular second moment, CON is contrast, ENT is entropy, HOM is homogeneity, L represents the number of image gray levels, p (i, j) is an element of a gray level co-occurrence matrix, and i, j are row elements and column elements of the matrix elements, respectively.
According to the method, R, G, B spectrum mean values on three wave bands are selected as the spectrum characteristic values of the image object, the difference of the spectrum characteristics of the image object at different wave bands is comprehensively considered, and the extracted spectrum characteristic values are guaranteed to be representative; according to the method, four most commonly used texture features, namely Angular Second Moment (ASM), Contrast (CON), Entropy (ENT) and Homogeneity (HOM), are selected as texture feature values, redundancy among the texture feature values is avoided, and the selected texture feature values can reflect texture differences among different ground objects.
And 3, screening a sample: randomly selecting a certain number of grid images from grid images corresponding to various places, carrying out superpixel segmentation, and taking the obtained superpixels as initial samples of the various places; because the proportion of the changed areas in the image is small, and most areas are unchanged, based on the unchanged information in the existing data result, namely the homogeneity characteristic values of unchanged ground samples are distributed and concentrated, the original samples with abnormal homogeneity are removed by a boxplot abnormal value analysis method to screen the experimental samples.
Randomly selecting a certain number of pattern spots from the grid images corresponding to each land class, obtaining the grid images of the pattern spots by using a mask extraction method, performing superpixel segmentation on the grid images of the pattern spots of each land class, and taking the obtained superpixels as initial samples of each land class;
calculating the homogeneity characteristic value of each type of initial sample, drawing a box line graph of the distribution of the homogeneity characteristic value, based on the characteristic that the distribution of the homogeneity characteristic value of the type of samples in an unchanged area in an image is concentrated, eliminating abnormal samples with the homogeneity characteristic value larger than an Upper Bound (UB) and smaller than a Lower Bound (LB) by using an anomaly analysis method, and selecting a required number of experimental samples from the rest samples.
The expressions of the upper bound UB and the lower bound LB of the box line graph are respectively as follows:
UB＝Qu+1.5IQR
LB＝Ql-1.5IQR
where Qu is the upper quartile of the boxplot distribution, Ql is the lower quartile, and IQR is the difference between the upper quartile Qu and the lower quartile Ql.
According to the method and the device, the optimal image characteristics are selected for different types of change detection objects, and the detection precision is improved.
The kernel function parameters of the model include C and γ, where C is a penalty factor, i.e., tolerance to errors. If C is too large or too small, the generalization ability of the SVM model is deteriorated. Gamma is a parameter of the RBF function itself after the function is selected as the kernel function. Implicitly determines the distribution of the data after mapping to a new feature space, wherein the larger the gamma, the fewer the support vectors, and the smaller the gamma value, the more the support vectors.
Under the guidance of the existing three-tone data, the method can realize automatic screening of the ground samples and solve the problem of sample library construction by using the method for analyzing the anomaly of the homogeneity characteristic value; and aiming at different change types, an object-oriented detection method with optimal image characteristics is selected, so that the change detection precision is effectively improved.
As a possible implementation manner of this embodiment, the method further includes the following steps:
step 6, evaluating the precision of the detection result; and comparing and analyzing the change detection result with a true value image visually interpreted, selecting 3 indexes of a virtual inspection rate, a missed inspection rate and a correct rate to evaluate the precision, and evaluating the accuracy of the change detection result.
The expressions of the virtual inspection rate, the missed inspection rate and the correct rate are as follows:
wherein FP is the number of pixels which are actually unchanged and detected as changed, FC is the total number of pixels which are detected as changed, FN is the number of pixels which are actually changed and detected as unchanged, PC is the total number of pixels which are actually changed, TP is the number of pixels which are actually changed and detected as changed, TN is the number of pixels which are actually unchanged and detected as unchanged, and N is the total number of all pixels in the detection area.
And comparing and analyzing the change detection result with a true value image visually interpreted, and selecting 3 indexes of a False Rate (FR), a Missing Rate (MR) and an Accuracy Rate (ACCURACY) for precision evaluation. The lower the false detection rate and the omission factor are, the higher the accuracy is, and the higher the precision of the detection result is.
Example two:
as shown in fig. 2, an arable land range change detection apparatus based on invariant information sample screening according to an embodiment of the present invention includes:
the classification information extraction module is used for analyzing and processing the homeland utilization survey data, acquiring grid images corresponding to all land types, and determining the acquired farmland range images as the to-be-detected area;
the spectrum and texture feature extraction module is used for performing super-pixel segmentation on the grid image of the region to be detected to obtain an image object and extracting the spectrum feature and the texture feature of the image object;
the sample screening module is used for randomly selecting a certain number of grid images from the grid images corresponding to all the places, carrying out superpixel segmentation, taking the obtained superpixels as initial samples of all the places, and screening experimental samples;
the optimal feature selection module is used for selecting different image features, performing classification experiments on different land samples obtained through automatic screening, and determining image feature differences between cultivated land and other different land types;
the change region detection module is used for training the SVM classification model by using an experimental sample and adaptively determining kernel function parameters of the model by adopting a grid search method; and selecting the optimal image characteristics according to different farmland change types, classifying the superpixel segmentation results of the to-be-detected region by using a trained SVM model, extracting changed superpixel objects, and detecting the farmland change region.
As a possible implementation manner of this embodiment, the classification information extraction module includes:
the remote sensing image acquisition module is used for acquiring a remote sensing image utilized by the state and the soil;
the system comprises a map spot acquisition module, a map spot classification module, a map image acquisition module and a map image classification module, wherein the map spot acquisition module is used for acquiring a vector range of each land by processing a remote sensing image by using a vector boundary and a category attribute of a map spot;
and the to-be-detected region determining module is used for performing mask extraction on the remote sensing images of the obtained vector ranges of all the land types to obtain raster images corresponding to all the land types, and determining the range of the to-be-detected region according to the obtained cultivated land range images.
As a possible implementation manner of this embodiment, the spectrum and texture feature extraction module includes:
a spectrum characteristic value extraction module for extracting a spectrum average value I of the image object in R, G, B three bandsR、IGAnd IBAs a spectral feature value;
and the texture characteristic value extraction module is used for calculating the texture characteristic value of the image object by utilizing the gray level co-occurrence matrix, wherein the texture characteristic value comprises an angle second moment, contrast, entropy and homogeneity.
As a possible implementation manner of this embodiment, the sample screening module includes:
the initial sample acquisition module is used for randomly selecting a certain number of patches from the grid images corresponding to all the land types, acquiring the grid images of the patches by using a mask extraction method, performing superpixel segmentation on the grid images of the patches of all the land types, and taking the obtained superpixels as initial samples of all the land types;
the experimental sample screening module is used for calculating homogeneity characteristic values of various regional initial samples, drawing a box line graph of homogeneity characteristic value distribution, based on the characteristic that the homogeneity characteristic distribution of the regional samples in an image is concentrated, eliminating abnormal samples with the homogeneity characteristic values larger than an upper limit and smaller than a lower limit by using an abnormality analysis method, and selecting the required number of experimental samples from the rest samples. As a possible implementation manner of this embodiment, the apparatus further includes:
and the detection result evaluation module is used for comparing and analyzing the change detection result with the visually-interpreted true value image, selecting 3 indexes of the virtual inspection rate, the omission factor and the accuracy rate to perform precision evaluation, and evaluating the accuracy of the change detection result.
The cultivated land range change detection device based on invariant information sample screening according to the embodiment is used for detecting the change of a high-resolution remote sensing image in a certain county within a cultivated land range, and as shown in fig. 3, the specific detection steps are as follows:
step a, extracting classification information; the method comprises the steps of obtaining homeland utilization survey data from a three-tone fruit database in the county, obtaining vector ranges of cultivated land and other land types by utilizing vector boundaries and category attributes of map spots of the land types, obtaining grid images of the cultivated land and other land types by adopting a mask extraction mode through original images as shown in fig. 4(a), and obtaining images of the cultivated land ranges as shown in fig. 4 (b). The method mainly aims to quickly position a research area to an arable land range to be detected and simultaneously facilitate subsequent selection of samples in various land vector ranges.
B, extracting spectral and textural features; and (4) performing superpixel segmentation on the cultivated land range image to be detected to obtain a superpixel object, wherein the superpixel segmentation result is shown in fig. 4(c), and extracting the spectral and texture characteristic values of the superpixel object. The spectral mean values of R, G, B three bands are selected as the spectral characteristic values of the image object, and the main purpose is to comprehensively consider the differences of the spectral characteristics of the image object at different bands and ensure that the extracted spectral characteristic values are representative; four most common texture features of angular second moment, contrast, entropy and homogeneity are selected as texture feature values, so that redundancy among the texture feature values is avoided, and the selected texture feature values can reflect texture differences among different ground objects.
Step c, automatically screening samples; randomly selecting a certain number of patches from the vector image layers of each land category, obtaining a grid image of the patches by using a mask extraction method, performing superpixel segmentation on the grid image of the patches of each land category, and taking the obtained superpixel as an initial sample of each land category. Calculating the homogeneity characteristic value of each type of initial sample, drawing a box line graph of homogeneity characteristic value distribution, based on the characteristic that the homogeneity characteristic distribution of the type of samples in the unchanged area in the image is concentrated, eliminating the samples with abnormal homogeneity characteristic values by using an abnormal analysis method, and selecting the experimental samples with the quantity required by the experiment from the rest samples. Because the proportion of the changed areas in the image is small, and most areas are unchanged, based on the unchanged information in the existing data result, namely the homogeneity characteristic values of unchanged ground samples are distributed and concentrated, the original samples with abnormal homogeneity are removed by a boxplot abnormal value analysis method to screen the experimental samples. The experimental samples were prepared as follows 7: the ratio of 3 is divided into a training set and a test set.
Step d, selecting the optimal characteristics; selecting different image characteristics, obtaining samples of cultivated land and other land types through automatic screening to carry out classification experiments, and determining image characteristic differences between the cultivated land and the different land types through precision evaluation results of the classification experiments. The land classification experiment proves that the precision of distinguishing cultivated land and construction land by spectral features is highest, and the precision of distinguishing cultivated land, garden land and forest land by textural features is highest, so that the optimal image feature needs to be selected according to different cultivated land change types, and the detection precision is improved.
Step e, detecting a change area; automatically screening cultivated land and other land samples, training an SVM classification model by using training set samples, and adaptively determining the optimal kernel function parameters C and gamma of the model by adopting a grid search method; and selecting optimal image characteristics according to different change types, classifying the super-pixel segmentation results of the region to be detected by using a trained SVM model, and extracting changed super-pixel objects so as to detect the farmland change region.
F, evaluating the precision of the detection result; in order to evaluate the Accuracy of the change detection result, the change detection result of the technical scheme is compared with a true value image which is visually interpreted, and 3 indexes of a False detection Rate (FR), a Missing detection Rate (MR) and an Accuracy Rate (ACCURACY) are selected for precision evaluation.
In order to verify the effectiveness of the invention, two groups of comparison experiments are set, wherein the method comprises the steps of screening a sample without data guide and directly and randomly selecting the sample for experiment, and the method comprises the steps of selecting image characteristics without data guide and directly utilizing spectrum and texture characteristics for experiment. The results of the evaluation of the detection accuracy by the different methods are shown in table 1:
TABLE 1 comparison of evaluation results of detection accuracy by different methods
And (3) analyzing the detection precision evaluation result: in the aspect of false detection rate, the method of the invention reduces 15.54 percent and 38.15 percent respectively compared with the method one and the method two; in the aspect of omission rate, the method of the invention reduces 4.4 percent and 0.86 percent respectively compared with the first method and the second method; in the aspect of accuracy, the method of the invention is improved by 2.39% and 8.06% compared with the first method and the second method respectively. Therefore, the method can effectively reduce the false detection rate and the omission factor under the guidance of the existing data, and accurately detect the farmland change area.
Example three: the arable land range change detection equipment based on invariant information sample screening provided by the embodiment of the invention comprises a processor and a memory, wherein computer program instructions are stored in the memory;
the processor reads and executes the computer program instructions to implement the arable land range change detection method based on invariant information sample screening as described in any of the above.
Specifically, the memory and the processor can be general-purpose memory and processor, which are not limited in particular, and the processor can execute the method for detecting farmland range change based on invariant information sample screening when the processor runs the computer program stored in the memory.
Those skilled in the art will appreciate that the configuration of the agrarian range change detection apparatus based on the invariant information sample screening does not constitute a limitation of the detection apparatus and may include more or fewer components than illustrated, or combine certain components, or split certain components, or a different arrangement of components.
In some embodiments, the detection device may further include a touch screen operable to display a graphical user interface (e.g., a launch interface for an application) and receive user operations with respect to the graphical user interface (e.g., launch operations with respect to the application). A particular touch screen may include a display panel and a touch panel. The Display panel may be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), and the like. The touch panel may collect contact or non-contact operations on or near the touch panel by a user and generate preset operation instructions, for example, operations of the user on or near the touch panel using any suitable object or accessory such as a finger, a stylus, etc. In addition, the touch panel may include two parts of a touch detection device and a touch controller. The touch detection device detects the touch direction and gesture of a user, detects signals brought by touch operation and transmits the signals to the touch controller; the touch controller receives touch information from the touch detection device, converts the touch information into information capable of being processed by the processor, sends the information to the processor, and receives and executes commands sent by the processor. In addition, the touch panel may be implemented by various types such as a resistive type, a capacitive type, an infrared ray, a surface acoustic wave, and the like, and may also be implemented by any technology developed in the future. Further, the touch panel may overlay the display panel, a user may operate on or near the touch panel overlaid on the display panel according to a graphical user interface displayed by the display panel, the touch panel detects an operation thereon or nearby and transmits the operation to the processor to determine a user input, and the processor then provides a corresponding visual output on the display panel in response to the user input. In addition, the touch panel and the display panel can be realized as two independent components or can be integrated.
Example four:
corresponding to the method for starting the application program in the embodiment 3, the embodiment of the present invention further provides a storage medium, wherein the storage medium stores a computer program, and the computer program is executed by a processor to execute the steps of any cultivated land area change detection method based on invariant information sample screening as described above.
The starting device of the application program provided by the embodiment of the application program can be specific hardware on the device or software or firmware installed on the device. The device provided by the embodiment of the present application has the same implementation principle and technical effect as the foregoing method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the foregoing method embodiments where no part of the device embodiments is mentioned. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the foregoing systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, 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.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, a division of modules is merely a division of logical functions, and an actual implementation may have another division, and for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or modules through some communication interfaces, and may be in an electrical, mechanical or other form.
Modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments provided in the present application may be integrated into one processing module, or each module may exist alone physically, or two or more modules are integrated into one module.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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 processor, 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.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (10)
1. A farmland range change detection method based on invariant information sample screening is characterized by comprising the following steps:
extracting classification information: analyzing and processing the homeland by utilizing survey data, acquiring grid images corresponding to all land types, and determining the acquired farmland range images as the to-be-detected area;
extracting spectral and textural features: performing super-pixel segmentation on the grid image of the region to be detected to obtain an image object, and extracting the spectral feature and the texture feature of the image object;
screening samples: randomly selecting a certain number of grid images from grid images corresponding to various places, carrying out superpixel segmentation, and taking the obtained superpixels as initial samples of the various places; based on invariant information in the existing data results, an initial sample with abnormal homogeneity is removed by a boxplot abnormal value analysis method, and an experimental sample is obtained through screening;
selecting the optimal characteristics: selecting different image characteristics, performing classification experiments on different land samples obtained through automatic screening, and determining image characteristic differences between cultivated land and other land types;
detecting a change area: training an SVM classification model by using an experimental sample, and adaptively determining kernel function parameters of the model by adopting a grid search method; and selecting the optimal image characteristics according to different farmland change types, classifying the superpixel segmentation results of the to-be-detected region by using a trained SVM model, extracting changed superpixel objects, and detecting the farmland change region.
2. The cultivated land range change detection method based on invariant information sample screening of claim 1, wherein: the extracting the classification information comprises the following steps:
obtaining a remote sensing image utilized by the homeland;
processing the remote sensing image by using the vector boundary and the category attribute of the image spot to obtain the vector range of each land category;
and performing mask extraction on the obtained remote sensing images of the vector ranges of all the land types to obtain grid images corresponding to all the land types, and determining the range of the area to be detected according to the obtained cultivated land range images.
3. The cultivated land range change detection method based on invariant information sample screening of claim 1, wherein: the extraction of the spectral and textural features comprises the following steps:
extracting the spectral average value I of the image object in R, G, B three bandsR、IGAnd IBAs a spectral feature value;
calculating a texture characteristic value of the image object by utilizing the gray level co-occurrence matrix;
the texture feature values include angular second moment, contrast, entropy and homogeneity.
4. The cultivated land range change detection method based on invariant information sample screening of claim 3 wherein: the spectral mean value IR、IGAnd IBThe expression of (a) is:
wherein n is the number of pixels in each image object, Ii R、Ii G、Ii BRespectively is the spectral value of the ith pixel in the super pixel in R, G, B three bands;
the expression of the texture feature value is as follows:
wherein ASM is angular second moment, CON is contrast, ENT is entropy, HOM is homogeneity, L represents the number of image gray levels, p (i, j) is an element of a gray level co-occurrence matrix, and i, j are row elements and column elements of the matrix elements, respectively.
5. The cultivated land range change detection method based on invariant information sample screening of claim 1, wherein: the screening sample comprises:
randomly selecting a certain number of pattern spots from the grid images corresponding to each land class, obtaining the grid images of the pattern spots by using a mask extraction method, performing superpixel segmentation on the grid images of the pattern spots of each land class, and taking the obtained superpixels as initial samples of each land class;
calculating the homogeneity characteristic value of each type of initial sample, drawing a box line graph of the distribution of the homogeneity characteristic value, based on the characteristic that the homogeneity characteristic distribution of the type of samples in an unchanged area in an image is concentrated, eliminating abnormal samples with the homogeneity characteristic value larger than an upper limit and smaller than a lower limit by using an abnormal analysis method, and selecting a required number of experimental samples from the rest samples.
6. The cultivated land range change detection method based on invariant information sample screening according to any one of claims 1 to 5, characterized in that: also comprises the following steps:
evaluating the precision of the detection result; and comparing and analyzing the change detection result with a true value image visually interpreted, selecting 3 indexes of a virtual inspection rate, a missed inspection rate and a correct rate to evaluate the precision, and evaluating the accuracy of the change detection result.
7. The utility model provides a arable land scope change detection device based on unchangeable information sample screening which characterized in that: the method comprises the following steps:
the classification information extraction module is used for analyzing and processing the homeland utilization survey data, acquiring grid images corresponding to all land types, and taking the acquired farmland range images as the to-be-detected area;
the spectrum and texture feature extraction module is used for performing super-pixel segmentation on the grid image of the region to be detected to obtain an image object and extracting the spectrum feature and the texture feature of the image object;
the sample screening module is used for randomly selecting a certain number of grid images from the grid images corresponding to all the places and carrying out superpixel segmentation, and the obtained superpixels are used as initial samples of all the places; based on invariant information in the existing data results, an initial sample with abnormal homogeneity is removed by a boxplot abnormal value analysis method, and an experimental sample is obtained through screening;
the optimal feature selection module is used for selecting different image features, performing classification experiments on different land samples obtained through automatic screening, and determining image feature differences between cultivated land and other land types; the change region detection module is used for training the SVM classification model by using an experimental sample and adaptively determining kernel function parameters of the model by adopting a grid search method; and selecting the optimal image characteristics according to different farmland change types, classifying the superpixel segmentation results of the to-be-detected region by using a trained SVM model, extracting changed superpixel objects, and detecting the farmland change region.
8. The cultivated land extent change detection apparatus based on invariable information sample screening of claim 7 wherein: further comprising:
and the detection result evaluation module is used for comparing and analyzing the change detection result with the visually-interpreted true value image, selecting 3 indexes of the virtual inspection rate, the omission factor and the accuracy rate to perform precision evaluation, and evaluating the accuracy of the change detection result.
9. The utility model provides a arable land scope change check out test set based on unchangeable information sample screening which characterized in that: comprising a processor, and a memory storing computer program instructions;
the processor reads and executes the computer program instructions to implement the arable land range change detection method based on invariant information sample screening according to any one of claims 1-4.
10. A storage medium, characterized by: the storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for detecting a tillage range change based on invariant information sample screening as claimed in any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111184439.0A CN113989657B (en) | 2021-10-11 | 2021-10-11 | Cultivated land range change detection method and device based on invariable information sample screening |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111184439.0A CN113989657B (en) | 2021-10-11 | 2021-10-11 | Cultivated land range change detection method and device based on invariable information sample screening |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113989657A true CN113989657A (en) | 2022-01-28 |
| CN113989657B CN113989657B (en) | 2024-05-10 |
Family
ID=79738174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111184439.0A Active CN113989657B (en) | 2021-10-11 | 2021-10-11 | Cultivated land range change detection method and device based on invariable information sample screening |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113989657B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115239682A (en) * | 2022-08-03 | 2022-10-25 | 国网江苏省电力有限公司无锡供电分公司 | Cable line external damage prevention monitoring and alarming method and system |
| CN115690591A (en) * | 2023-01-05 | 2023-02-03 | 速度时空信息科技股份有限公司 | Remote sensing image farmland non-agricultural change detection method based on deep learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846832A (en) * | 2018-05-30 | 2018-11-20 | 理大产学研基地(深圳)有限公司 | A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data |
| CN109344810A (en) * | 2018-11-26 | 2019-02-15 | 国智恒北斗科技集团股份有限公司 | A kind of arable land use change monitoring method and system based on high score satellite remote sensing date |
| WO2020015326A1 (en) * | 2018-07-19 | 2020-01-23 | 山东科技大学 | Remote sensing image cloud shadow detection method supported by earth surface type data |
| WO2020063461A1 (en) * | 2018-09-30 | 2020-04-02 | 广州地理研究所 | Urban extent extraction method and apparatus based on random forest classification algorithm, and electronic device |
| CN111598045A (en) * | 2020-05-27 | 2020-08-28 | 中国科学院地理科学与资源研究所 | Remote sensing farmland change detection method based on object spectrum and mixed spectrum |
| CN111881801A (en) * | 2020-07-22 | 2020-11-03 | 中国测绘科学研究院 | Newly-added construction land remote sensing monitoring method and equipment based on invariant detection strategy |
| CN113269037A (en) * | 2021-04-14 | 2021-08-17 | 中南大学 | Single-phase image earth surface coverage change detection automation method |
-
2021
- 2021-10-11 CN CN202111184439.0A patent/CN113989657B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846832A (en) * | 2018-05-30 | 2018-11-20 | 理大产学研基地(深圳)有限公司 | A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data |
| WO2020015326A1 (en) * | 2018-07-19 | 2020-01-23 | 山东科技大学 | Remote sensing image cloud shadow detection method supported by earth surface type data |
| WO2020063461A1 (en) * | 2018-09-30 | 2020-04-02 | 广州地理研究所 | Urban extent extraction method and apparatus based on random forest classification algorithm, and electronic device |
| CN109344810A (en) * | 2018-11-26 | 2019-02-15 | 国智恒北斗科技集团股份有限公司 | A kind of arable land use change monitoring method and system based on high score satellite remote sensing date |
| CN111598045A (en) * | 2020-05-27 | 2020-08-28 | 中国科学院地理科学与资源研究所 | Remote sensing farmland change detection method based on object spectrum and mixed spectrum |
| CN111881801A (en) * | 2020-07-22 | 2020-11-03 | 中国测绘科学研究院 | Newly-added construction land remote sensing monitoring method and equipment based on invariant detection strategy |
| CN113269037A (en) * | 2021-04-14 | 2021-08-17 | 中南大学 | Single-phase image earth surface coverage change detection automation method |
Non-Patent Citations (2)
| Title |
|---|
| 冯文卿;眭海刚;涂继辉;孙开敏;黄伟明;: "高分辨率遥感影像的随机森林变化检测方法", 测绘学报, no. 11, 15 November 2017 (2017-11-15), pages 34 - 35 * |
| 魏东升;周晓光;: "遥感影像变化检测样本自动抽样", 遥感学报, no. 03, 25 May 2019 (2019-05-25), pages 110 - 121 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115239682A (en) * | 2022-08-03 | 2022-10-25 | 国网江苏省电力有限公司无锡供电分公司 | Cable line external damage prevention monitoring and alarming method and system |
| CN115239682B (en) * | 2022-08-03 | 2024-01-30 | 国网江苏省电力有限公司无锡供电分公司 | Cable line external damage prevention monitoring and alarming method and system |
| CN115690591A (en) * | 2023-01-05 | 2023-02-03 | 速度时空信息科技股份有限公司 | Remote sensing image farmland non-agricultural change detection method based on deep learning |
| CN115690591B (en) * | 2023-01-05 | 2023-03-31 | 速度时空信息科技股份有限公司 | Remote sensing image farmland non-agricultural change detection method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113989657B (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3408825B1 (en) | System and method for earth observation and analysis | |
| Zhang et al. | Automated robust crop-row detection in maize fields based on position clustering algorithm and shortest path method | |
| White et al. | Confirmation of post-harvest spectral recovery from Landsat time series using measures of forest cover and height derived from airborne laser scanning data | |
| CN108846832B (en) | Multi-temporal remote sensing image and GIS data based change detection method and system | |
| Mack et al. | A semi-automated approach for the generation of a new land use and land cover product for Germany based on Landsat time-series and Lucas in-situ data | |
| Riemann et al. | An effective assessment protocol for continuous geospatial datasets of forest characteristics using USFS Forest Inventory and Analysis (FIA) data | |
| CN113989657B (en) | Cultivated land range change detection method and device based on invariable information sample screening | |
| Jiang et al. | Quantitative analysis of cotton canopy size in field conditions using a consumer-grade RGB-D camera | |
| Wang et al. | Major forest increase on the Loess Plateau, China (2001–2016) | |
| CN112164062A (en) | Wasteland information extraction method and device based on remote sensing time sequence analysis | |
| Rich et al. | Detecting wind disturbance severity and canopy heterogeneity in boreal forest by coupling high-spatial resolution satellite imagery and field data | |
| CN107239794A (en) | Point cloud data segmentation method and terminal | |
| CN102930279B (en) | For the image-recognizing method that product quantity detects | |
| JP2014100099A (en) | Forest information management apparatus | |
| EP4081959A1 (en) | Agricultural decision support system | |
| Creasy et al. | Potential for individual tree monitoring in ponderosa pine dominated forests using unmanned aerial system structure from motion point clouds | |
| Liu et al. | Automated classification of stems and leaves of potted plants based on point cloud data | |
| CN111612868A (en) | Map optimization method and device | |
| Oehmcke et al. | Deep point cloud regression for above-ground forest biomass estimation from airborne LiDAR | |
| Gómez et al. | Integrated object-based spatiotemporal characterization of forest change from an annual time series of Landsat image composites | |
| CN111144340A (en) | Method and system for automatically monitoring human activities in natural reserve area based on night light and high-resolution remote sensing image | |
| CN118230184A (en) | Automatic identification system and method for distribution network line tree barriers based on satellite remote sensing | |
| Galli et al. | A novel way to validate UAS-based high-throughput phenotyping protocols using in silico experiments for plant breeding purposes | |
| Deng et al. | Extraction of farmland shelterbelts from remote sensing imagery based on a belt-oriented method | |
| Kimambo et al. | Using Landsat and Sentinel-2 spectral time series to detect East African small woodlots |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |