CN111832480A - Remote sensing identification method for rape planting area based on spectral characteristics - Google Patents

Remote sensing identification method for rape planting area based on spectral characteristics Download PDF

Info

Publication number
CN111832480A
CN111832480A CN202010672490.5A CN202010672490A CN111832480A CN 111832480 A CN111832480 A CN 111832480A CN 202010672490 A CN202010672490 A CN 202010672490A CN 111832480 A CN111832480 A CN 111832480A
Authority
CN
China
Prior art keywords
rape
period
classification
remote sensing
image
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
Application number
CN202010672490.5A
Other languages
Chinese (zh)
Other versions
CN111832480B (en
Inventor
张朝
韩继冲
骆玉川
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Normal University
Original Assignee
Beijing Normal University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Normal University filed Critical Beijing Normal University
Priority to CN202010672490.5A priority Critical patent/CN111832480B/en
Publication of CN111832480A publication Critical patent/CN111832480A/en
Application granted granted Critical
Publication of CN111832480B publication Critical patent/CN111832480B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06V20/188—Vegetation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/143—Sensing or illuminating at different wavelengths

Abstract

The invention discloses a rape planting area optical remote sensing identification technology based on spectral characteristics, which comprises the following steps: s1, defining the time period from the initial flowering time to about 25 days after the flower falling date of the rape as a proper rape classification period; s2: preprocessing Landsat series satellite reflectivity data of a research area in a classification period, and removing a cloud area in an image; s3: calculating three spectral indexes (NDVI, mNDVI and NRFIG) of all images in the classification period according to the reflectivity data; s4: respectively setting corresponding threshold values for the three spectral indexes, and segmenting and recombining the three spectral indexes to obtain rape classification maps at different time points; s5: and overlapping the rape classification results of all time points in the classification period to obtain a final rape identification area. The method is simple in calculation, does not need a large number of training samples, reduces the dependence on the availability of high-quality images of rape in the flowering phase, and provides a reliable method.

Description

Remote sensing identification method for rape planting area based on spectral characteristics
Technical Field
The invention relates to the technical field of agricultural remote sensing, in particular to a remote sensing identification method for rape planting areas based on spectral characteristics.
Background
Rape is a main economic crop and oil source in the world, wherein China is one of the countries with the largest rape planting area and the highest yield. The identification of the rape planting area is the basis for monitoring the growth vigor of the rape, predicting the rape yield and making relevant agricultural measures, and has very important significance for the import and export of the rape and the like.
The current main rape optical remote sensing identification methods are divided into a remote sensing image visual interpretation method combined with field investigation statistics, a machine learning classification method, a threshold classification method according to the feature of the phenological features and a rape identification method based on HSV color conversion. The field survey statistics is based on field measurement of relevant personnel, and the remote sensing image visual interpretation is mainly obtained by utilizing the unique visual characteristics (such as yellow green true color) of the rape on the flowering image to carry out artificial vectorization; the machine learning classification is mainly based on a large amount of real sample data, and the images of the rape flowering phase are classified by utilizing a machine learning algorithm (such as a support vector machine and an artificial neural network); the classification method for dividing the threshold value according to the phenological characteristics mainly sets a corresponding threshold value for extracting a planting area by utilizing the difference between a phenological curve (such as NDVI) of other vegetation types in the growth process of rape; the rape identification method based on HSV color transformation mainly converts the reflectivity data of the flowering phase into color characteristic data (color, purity and brightness) to identify a rape planting area.
The method based on field survey statistics and remote sensing image visual interpretation has high precision, but needs a great amount of manpower, financial resources and material resources. In addition, the basis for visual interpretation is the full satellite image available at the full flowering stage of rape in the target area. Supervised classification in machine learning requires a large amount of real training sample data, and images of the flowering phase are basic data for classification. The classification method for dividing the threshold value according to the feature of the phenological features does not need a large amount of sample data, and the method requires a continuously available high-quality image in the growing period of the rape, however, most areas cannot meet the premise. The HSV color conversion based method is based on the principle that the rape blossom is characterized by the color, so the method is premised on that a high-quality image is available in the flowering period. Therefore, the common defect of the current rape identification research is high dependence on the availability of flowering phase images, which limits the large-scale and long-time sequence research to a certain extent.
Accordingly, new techniques are needed to at least partially address the limitations present in the prior art.
Disclosure of Invention
In order to overcome the defect that the existing rape planting area optical remote sensing identification method is highly dependent on rape flowering phase images, the invention creates a new technical method capable of reducing the influence of flowering phase image quality on rape identification precision based on the phenological and spectral characteristics of rape.
According to one aspect of the invention, a technology for identifying a rape planting area by optical remote sensing based on spectral characteristics is provided, which comprises the following steps:
s1, determining the rape classification period of the target area, wherein the rape classification period comprises a rape flowering period and a late flowering period, the rape flowering period is a time period from the initial flowering date to the initial flower falling date of the rape, and the late flowering period is a time period of 20-30 days after the initial flower falling date, and the time period can be selected as the case may be, for example, 25 days;
s2, preprocessing the satellite remote sensing reflectivity images of different time points in the classification period of the target area, and removing the cloud-covered pixels in the reflectivity images to obtain basic reflectivity images;
s3, based on the basic reflectivity image of the step S2, the different wave band reflectivities are combined and calculated to obtain three spectral indexes including normalized vegetation index (NDVI), normalized water body index (mNDVI) and NRFIG at different time points in the classification period, wherein,
Figure BDA0002582826220000021
wherein B isgreeIs a green band reflectance, Bswir2Is the reflectivity of short wave infrared 2 wave bands;
s4: setting corresponding threshold values T for NDVI, mNDBI and NRFIG respectivelyVI，TWIAnd TFIWhen a certain pixel simultaneously satisfies the condition in the following formula (4) at a certain time point, the pixel at the time point is classified as rape:
Figure BDA0002582826220000031
wherein i represents the time of the satellite capturing the image, NDVIi，mNDWIi，NRFIgiRespectively representing the values of NDVI, mNDWI and NRFIg at time i; clasis is a classification result obtained based on the image of the i time;
s5: and (4) overlapping and combining the rape identification result image layers at different time points obtained in the S4, and taking a union set of all the image layers, namely the maximum value of the image elements at the same position on different image layers as a final spatial distribution map of the rape planting area, thereby obtaining the rape planting area.
According to an embodiment of the invention, the satellite is a Landsat satellite, or other suitable medium-low resolution satellite.
According to an embodiment of the present invention, the threshold value TVI0.50-0.60, TWI0.03-0.07, TFIRespectively 0.07-0.13. E.g. TVIIs 0.55, TWIIs 0.05, TFIIs 0.10. The skilled person can also make appropriate adjustments to the threshold value as the case may be.
According to an embodiment of the present invention, the codes for the initial flowering date and the initial flower fall date of Brassica napus can be determined according to the International plant growth coding System BBCH-scale (canola), the code for the initial flowering date being 60 and the code for the initial flower fall date being 69.
According to an embodiment of the present invention, the method for removing the pixel covered by the cloud in the image may be a CFMASK algorithm, or other suitable algorithms.
According to an embodiment of the invention, wherein:
Figure BDA0002582826220000041
Figure BDA0002582826220000042
in the formula, BgreenIs a green band reflectance, BredIs a red band reflectivity, BnirIs the reflectivity of the near infrared band, Bswir1Is short wave infrared 1 band reflectivity.
Compared with the prior art, the method does not need a large amount of ground actual sample data for training the classification model, and reduces the dependence on the rape flowering phase image availability. The method of the invention creatively provides a new index, and combines various indexes for processing, the operation is simple, no complex calculation formula is needed, but the method can identify the planting area of the rape with large range and high precision. The method provides support for researches on long-time rape planting area statistics under the pixel scale, the temporal and spatial change rule, yield prediction and the like.
Drawings
The same reference numbers in the drawings identify the same or similar elements or components. The objects and features of the present invention will become more apparent in view of the following description taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a schematic flow chart of the spectral feature-based rape optical remote sensing identification method of the invention;
FIG. 2 is a graph of three spectral indices during spring rape growth according to one embodiment of the present invention;
FIG. 3 is a graph of the results of a spring rape classification based on images of flowering and non-flowering times for a localized area according to one embodiment of the present invention;
FIG. 4 is a graph of the final recognition of the target area for the oil dish according to one embodiment of the present invention;
Detailed Description
In order to clearly illustrate the aspects of the present invention, preferred embodiments are given below, the following description being merely exemplary in nature and not intended to limit the application or uses of the present disclosure.
The specific application of the method of the present invention will be exemplified below with reference to the accompanying drawings, taking the source county of Qinghai province as a research target area.
FIG. 1 is a schematic flow chart of rape optical remote sensing identification based on spectral characteristics, which specifically comprises the following steps:
s1: selecting a Menyuan county of Qinghai province as a target area, wherein the area is a main production area of the spring rape. The range is between 100.86 DEG E and 102.59 DEG E, and between 37.06 DEG N and 37.88 DEG N. In 2017, the initial flowering time of the spring rapes in the region is about 7 months and 5 th, the flower falling time is about 7 months and 25 th, and the flower falling time is delayed by 25 days, namely, 7 months and 5-8 months and 20 th are defined as proper classification periods. In general, flowering and flowering times of oilseed rape can be determined from annual agricultural statistics in the area under study.
S2: acquiring reflectivity data of all landssat 8 satellites covering a target area in a classification period of 2017, wherein the spatial resolution of the data is 30m, and removing the cloud area in the image by using a CFMASK algorithm.
S3: based on the reflectivity data of different time points processed in the research period, three spectral indexes are respectively calculated to obtain raster image data of NDVI, mNDVI and NRFIG, wherein NDVI is a normalized vegetation index, mNDVI is a normalized water body index, and NRFIG is a spectral index created in the invention. Researches show that the three indexes are combined and processed to a certain extent, so that the high-precision identification of the rape planting area can be realized.
More specifically, the calculation of the three spectral indices can be seen in equations (1) - (3). The complete phenological curve of the three spectral indexes of the spring rape in the target area is shown in the attached figure 2, wherein the gray area is the flowering phase of the rape, and the dotted line range is the rape classification phase.
Figure BDA0002582826220000061
Figure BDA0002582826220000062
Figure BDA0002582826220000063
In the formula, BgreenIs a green band, BredIn the red band, BnirIn the near infrared band, Bswir1Is short wave infrared 1 band, Bswir2Is short wave infrared 2 wave band.
S4: and dividing the NDVI, the mNDVI and the NRFIG at different time points in the classification period into binary raster images according to a threshold value. More specifically, the NDVI, mNDBI and NRFIg are respectively set with corresponding threshold values TVI，TWIAnd TFISaid threshold value TVIMay be 0.50-0.60, TWIMay be 0.03-0.07, TFIMay be 0.07-0.13. It is understood that one skilled in the art can select an appropriate threshold value based on the teachings of the present invention and in conjunction with the particular situation.
In this embodiment, TVISelection was 0.55, TWIIs 0.05, TFIIs 0.10. Pixels with NDVI values greater than the threshold value of 0.55 can be assigned as 1, and the rest are assigned as 0; pixels with mNDBI values smaller than the threshold value of 0.05 can be assigned as 1, and the rest are assigned as 0; pixels with NRFIg greater than a threshold of 0.10 can be assigned a value of 1, with the remaining values assigned 0; where 1 indicates that the condition is satisfied and 0 indicates that the condition is not satisfied. And adding the results of the three spectral index divisions, wherein if the pixel value is 3, the area which simultaneously meets the three conditions (formula 4) is the identified rape planting area. According to the obtained rape planting area identification results at different time points, the pixel identified as the rape can be assigned with 1, and the other pixels can be assigned with 0. The image recognition results of the rape flowering phase and the local area in the late flowering phase are shown in the attached figure 3.
Figure BDA0002582826220000071
Wherein i represents the time of the Landsat8 satellite shooting image, classiNDVI, classification results obtained for i-time based imagesi，mNDWIi，NRFIgiNDVI, mNDVI and NRFIG at time i are shown, respectively.
S5: and (4) overlapping and combining the rape identification result image layers at different time points obtained in the step (S4), and taking a union set of all the image layers, namely the maximum value (formula 5) of the image elements at the same position in different image layers as a final spatial distribution map of the rape planting area, wherein the spatial resolution is 30m, and the time resolution is 1 year. The final rape planting area recognition result of the target area in 2017 is shown in fig. 4.
class＝max(classi) (5)
In order to verify the accuracy of the identification result of the rape planting area, four indexes of total precision, user precision, producer precision and F1 are used. Wherein the total precision represents the percentage of correctly classified samples to the total number of samples; the user precision of rape classification represents the percentage of the correctly classified rape samples in the total rape samples obtained by the classification method; the producer precision of the rape expresses the percentage of the correctly classified rape samples in the real total rape samples;
the F1 score is calculated as shown in equation (6).
Figure BDA0002582826220000081
Verification is performed based on the verification data set (338 rape sample points and 1199 non-rape sample points) visually interpreted by the flowering high-resolution images Sentinel 2 and Google Earth, and the obtained error matrix is shown in the following table 1:
TABLE 1
Figure BDA0002582826220000082
And calculating according to the error matrix to obtain the following precision of each index:
the total precision is (1174+382)/1587 × 100 ═ 98.0%;
the user precision of rape classification is 382/407 × 100 ═ 93.9%;
the producer precision of the rape classification is 382/388 × 100 ═ 98.5%;
f1 score 2 x (93.9% + 98.5%)/(93.9% + 98.5%) -96.1%
The result shows that the identification result of the method has higher precision and important scientific and economic values.
Compared with the prior art, the method does not need a large amount of ground actual sample data for training the classification model, can perform rape identification on the image of the non-flowering phase, and reduces the high dependence on the rape flowering phase image quality in the existing method. The method is simple to operate, does not need a complex calculation formula, and can identify the planting area of the rape in a large range with higher precision. The method provides support for researches on long-time rape planting area statistics under the pixel scale, the temporal and spatial change rule, yield prediction and the like.
The principles and embodiments of the present invention have been described herein using specific examples, which are presented solely to aid in the understanding of the apparatus and its core concepts; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims (6)

1. A remote sensing identification method for rape planting areas based on spectral characteristics comprises the following steps:
s1, determining the rape classification period of the target area, wherein the rape classification period comprises a rape flowering period and a late flowering period, the rape flowering period is a time period from the initial flowering date to the initial flower falling date of the rape, and the late flowering period is a time period of 20-30 days after the initial flower falling date;
s2, preprocessing the satellite remote sensing reflectivity images of different time points in the classification period of the target area, and removing the cloud-covered pixels in the reflectivity images to obtain basic reflectivity images;
s3, based on the basic reflectivity image of the step S2, the different wave band reflectivities are combined and calculated to obtain three spectral indexes including normalized vegetation index (NDVI), normalized water body index (mNDVI) and NRFIG at different time points in the classification period, wherein,
Figure FDA0002582826210000011
wherein B isgreeIs a green band reflectance, Bswir2Is the reflectivity of short wave infrared 2 wave bands;
s4: setting corresponding threshold values T for NDVI, mNDBI and NRFIG respectivelyVI，TWIAnd TFIWhen a certain pixel simultaneously satisfies the condition in the following formula (4) at a certain time point, the pixel at the time point is classified as rape:
Figure FDA0002582826210000012
wherein i represents the time of the satellite capturing the image, NDVIi，mNDWIi，NRFIgiRespectively representing the values of NDVI, mNDWI and NRFIg at time i; clasiThe classification result is obtained based on the image of the i time;
s5: and (4) overlapping and combining the rape identification result image layers at different time points obtained in the S4, and taking a union set of all the image layers, namely the maximum value of the image elements at the same position on different image layers as a final spatial distribution map of the rape planting area, thereby obtaining the rape planting area.
2. The remote sensing identification method for rape planting areas as claimed in claim 1, wherein the satellite is a Landsat satellite.
3. The remote sensing rape planting area identification method according to claim 1, wherein the threshold T is setVI0.50-0.60, TWI0.03-0.07, TFIIs 0.07-0.13.
4. The remote sensing method for rape planting area according to claim 1, wherein the codes of initial flowering date and initial flower falling date of rape are determined according to the international plant growth coding system BBCH-scale (canola), the code of initial flowering date is 60, and the code of initial flower falling date is 69.
5. The remote sensing identification method for rape planting areas according to claim 1, wherein the method for removing the pixel covered by the cloud in the image is a CFMASK algorithm.
6. The remote sensing identification method for regional rape planting areas of claim 1, wherein,
Figure FDA0002582826210000021
Figure FDA0002582826210000022
in the formula, BgreeIs a green band reflectance, BredIs a red band reflectivity, BnirIs the reflectivity of the near infrared band, Bswir1Is short wave infrared 1 band reflectivity.
CN202010672490.5A 2020-07-14 2020-07-14 Remote sensing identification method for rape planting area based on spectral characteristics Active CN111832480B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010672490.5A CN111832480B (en) 2020-07-14 2020-07-14 Remote sensing identification method for rape planting area based on spectral characteristics

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010672490.5A CN111832480B (en) 2020-07-14 2020-07-14 Remote sensing identification method for rape planting area based on spectral characteristics

Publications (2)

Publication Number Publication Date
CN111832480A true CN111832480A (en) 2020-10-27
CN111832480B CN111832480B (en) 2021-04-23

Family

ID=72922690

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010672490.5A Active CN111832480B (en) 2020-07-14 2020-07-14 Remote sensing identification method for rape planting area based on spectral characteristics

Country Status (1)

Country Link
CN (1) CN111832480B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112883987A (en) * 2021-03-11 2021-06-01 成都数之联科技有限公司 Target extraction method, system, device and medium based on remote sensing spectral characteristics
CN113158767A (en) * 2021-03-02 2021-07-23 山东科技大学 Spectral characteristic-based oncomelania breeding place identification method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107271382A (en) * 2017-06-02 2017-10-20 西北农林科技大学 A kind of different growing rape leaf SPAD value remote sensing estimation methods
CN107527014A (en) * 2017-07-20 2017-12-29 武汉珈和科技有限公司 Crops planting area RS statistics scheme of sample survey design method at county level
CN110175931A (en) * 2019-05-10 2019-08-27 北京师范大学 A kind of method of a wide range of rapidly extracting crop acreage and phenology information
CN110472184A (en) * 2019-08-22 2019-11-19 电子科技大学 A kind of cloudy misty rain area rice recognition methods based on Landsat remotely-sensed data
CN111209871A (en) * 2020-01-09 2020-05-29 河南大学 Rape planting land remote sensing automatic identification method based on optical satellite image

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107271382A (en) * 2017-06-02 2017-10-20 西北农林科技大学 A kind of different growing rape leaf SPAD value remote sensing estimation methods
CN107527014A (en) * 2017-07-20 2017-12-29 武汉珈和科技有限公司 Crops planting area RS statistics scheme of sample survey design method at county level
CN110175931A (en) * 2019-05-10 2019-08-27 北京师范大学 A kind of method of a wide range of rapidly extracting crop acreage and phenology information
CN110472184A (en) * 2019-08-22 2019-11-19 电子科技大学 A kind of cloudy misty rain area rice recognition methods based on Landsat remotely-sensed data
CN111209871A (en) * 2020-01-09 2020-05-29 河南大学 Rape planting land remote sensing automatic identification method based on optical satellite image

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
JIE SHAN等: "Study on Extraction Methods of Rape Area Based on GF-1 Satellite Image", 《2018 7TH INTERNATIONAL CONFERENCE ON AGRO-GEOINFORMATICS (AGRO-GEOINFORMATICS)》 *
佘宝等: "基于红边位置变化特征的油菜种植区域提取", 《农业工程学报》 *
刘明月等: "基于MODIS时序数据的Landsat8影像选取及面向对象分类方法的农作物分类", 《土壤与作物》 *
马尚杰等: "基于多时相环境星数据的甘蔗收割过程遥感监测", 《农业工程学报》 *
魏传文等: "基于多时相环境卫星的冬前油菜种植面积估算", 《中国农业信息》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113158767A (en) * 2021-03-02 2021-07-23 山东科技大学 Spectral characteristic-based oncomelania breeding place identification method
CN112883987A (en) * 2021-03-11 2021-06-01 成都数之联科技有限公司 Target extraction method, system, device and medium based on remote sensing spectral characteristics

Also Published As

Publication number Publication date
CN111832480B (en) 2021-04-23

Similar Documents

Publication Publication Date Title
CN111598019B (en) Crop type and planting mode identification method based on multi-source remote sensing data
CN112183209B (en) Regional crop classification method and system based on multidimensional feature fusion
CN110751019B (en) High-resolution image crop automatic extraction method and device based on deep learning
CN109146889A (en) A kind of field boundary extracting method based on high-resolution remote sensing image
CN108458978B (en) Sensitive waveband and waveband combination optimal tree species multispectral remote sensing identification method
CN113221765B (en) Vegetation phenological period extraction method based on digital camera image effective pixels
CN108710864B (en) Winter wheat remote sensing extraction method based on multi-dimensional identification and image noise reduction processing
CN111832480B (en) Remote sensing identification method for rape planting area based on spectral characteristics
CN112766155A (en) Deep learning-based mariculture area extraction method
CN116543316B (en) Method for identifying turf in paddy field by utilizing multi-time-phase high-resolution satellite image
CN114926748A (en) Soybean remote sensing identification method combining Sentinel-1/2 microwave and optical multispectral images
CN113033453A (en) Method and system suitable for remote sensing identification of crop types in landscape crushing area
CN110765977A (en) Method for extracting wheat lodging information based on multi-temporal remote sensing data of unmanned aerial vehicle
CN104951754A (en) Sophisticated crop classifying method based on combination of object oriented technology and NDVI (normalized difference vegetation index) time series
CN111008664A (en) Hyperspectral sea ice detection method based on space-spectrum combined characteristics
Olenskyj et al. End-to-end deep learning for directly estimating grape yield from ground-based imagery
CN116129260A (en) Forage grass image recognition method based on deep learning
CN111753834B (en) Planting land block structure semantic segmentation method and device based on deep neural network
Lin et al. Estimating litchi flower number using a multicolumn convolutional neural network based on a density map
Fan et al. An improved Deeplab based model for extracting cultivated land information from high definition remote sensing images
Rony et al. BottleNet18: Deep Learning-Based Bottle Gourd Leaf Disease Classification
CN115147730A (en) Remote sensing classification method combining full convolution neural network and ensemble learning
CN114612794A (en) Remote sensing identification method for land covering and planting structure in finely-divided agricultural area
CN114463642A (en) Cultivated land plot extraction method based on deep learning
CN115035423A (en) Hybrid rice male and female parent identification and extraction method based on unmanned aerial vehicle remote sensing image

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