Paddy field extraction method and device based on remote sensing cloud computing platform and phenological characteristics

Abstract

The application provides a paddy field extraction method and device based on a remote sensing cloud computing platform and a phenological feature, and the method comprises the following steps: determining a target feature of a pixel of a remote sensing data set based on remote sensing data sets of a to-be-identified region in different growth periods of crops; the target feature comprises a paddy field feature, a wetland feature and a farmland feature; stacking the remote sensing data set based on the target feature of the pixel to determine synthesized first image data; performing image segmentation on the first image data to determine second image data after segmentation; and in the case that the target feature of the pixel of the second image data meets a preset condition, extracting paddy field information of the to-be-identified region. The paddy field extraction method based on the remote sensing cloud computing platform and the phenological feature provided by the application realizes accurate and effective extraction of the paddy field by fully analyzing the differences in the phenology of different land cover types and converting the differences into feature parameters calculated by using the remote sensing data.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of agriculture and remote sensing technology, and in particular to a paddy field extraction method and device based on a remote sensing cloud computing platform and phenological characteristics. BACKGROUND

[0002] Due to the unique growth conditions and high yield of rice, paddy fields play an important role in water security, food security, climate change and human health, and it is crucial to quickly and accurately extract the spatial distribution of paddy fields in a region. However, due to the relatively complex underlying surface of a region, especially a large region, it is difficult to accurately and effectively extract paddy fields.

[0003] The rapid development of remote sensing technology, due to its continuous improvement in spectral, spatial and temporal resolution, has made it effective in the field of resource and environment, providing effective support for the extraction of large-area paddy fields. However, in order to facilitate the management and irrigation of paddy fields, they are often divided into many small grids, making it difficult for medium-resolution images to effectively distinguish paddy fields from other underlying surfaces.

[0004] In addition, it is also difficult to extract paddy fields using medium or high resolution remote sensing images. Due to the similarity of the spectrum, it is difficult to distinguish paddy fields from other surface cover types in terms of spectrum using single-scene remote sensing images. The phenological model mainly detects the flooding signal of rice in the transplanting period through time-series remote sensing data, but wetlands and floating plants have similar flooding signals to paddy fields in the transplanting period, making it difficult to effectively distinguish paddy fields from other surface cover types. SUMMARY

[0005] To solve the problems in the prior art, the present application provides a paddy field extraction method and device based on a remote sensing cloud computing platform and phenological characteristics.

[0006] The present application provides a paddy field extraction method based on a remote sensing cloud computing platform and phenological characteristics, comprising:

[0007] Based on the remote sensing data set of the crop in different growth periods in the region to be identified, the target characteristics of the pixels of the remote sensing data set are determined; the remote sensing data set includes: first remote sensing data in the transplanting period, second remote sensing data in the growth period and third remote sensing data in the harvesting period; the target characteristics include: paddy field characteristics, wetland characteristics and farmland characteristics;

[0008] Based on the target characteristics of the pixels, the remote sensing data set is stacked to determine the synthesized first image data;

[0009] The first image data is subjected to image segmentation to determine the second image data after segmentation;

[0010] In a case where the target feature of the pixel of the second image data meets a preset condition, paddy field information of the region to be identified is extracted.

[0011] In some embodiments, the determining of the target feature of the pixel of the remote sensing data set based on the remote sensing data sets of different growth periods of the crop comprises:

[0012] The first flooded pixel corresponding to the transplanting period is determined based on the band of the first pixel corresponding to the first remote sensing data.

[0013] The paddy field feature is determined based on the number of the first flooded pixel.

[0014] The second flooded pixel corresponding to the harvesting period is determined based on the band of the second pixel corresponding to the third remote sensing data.

[0015] The wetland feature is determined based on the number of the second flooded pixel.

[0016] The farmland feature is determined based on the normalized difference vegetation index of the second remote sensing data and the normalized difference vegetation index of the third remote sensing data.

[0017] In some embodiments, before the determining of the target feature of the pixel of the remote sensing data set based on the remote sensing data sets of different growth periods of the crop, the method further comprises:

[0018] The remote sensing data set is cloud-masked.

[0019] In some embodiments, the expression of the paddy field feature is as follows:

[0020]

[0021] Wherein, Paddy index represents the paddy field feature, F (Transplanting) represents the flooding frequency of the first pixel, ∑Total represents the total number of the first pixel observed by the satellite during the transplanting period, ∑Bad represents the number of invalid observations of the first pixel during the transplanting period, and ∑Flooded Pixel (Transplanting) represents the number of the first flooded pixel.

[0022] In some embodiments, the expression of the wetland feature is as follows:

[0023]

[0024] Wherein, Wetland index represents wetland characteristics, F(Harvesting) represents the frequency of flooding of the second pixel, ∑Total' represents the total number of observations of the second pixel during the harvest period, ∑Bad' represents the number of invalid observations of the second pixel during the harvest period, and ∑Flooded Pixel(Harvesting) represents the number of second flooded pixels.

[0025] In some embodiments, the expression of the cropland characteristics is as follows:

[0026]

[0027] Wherein, Cropland index represents cropland characteristics, NDVI(Growing) represents the maximum value of the normalized vegetation index during the growing period, and NDVI(Transplanting) represents the maximum value of the normalized vegetation index during the transplanting period.

[0028] The application also provides a paddy field extraction device based on a remote sensing cloud computing platform and phenological characteristics, comprising:

[0029] A first determination module is configured to determine target characteristics of pixels of a remote sensing data set based on remote sensing data sets of a to-be-identified region during different growth periods of crops; the remote sensing data set comprises first remote sensing data during a transplanting period, second remote sensing data during a growing period, and third remote sensing data during a harvest period; and the target characteristics comprise paddy field characteristics, wetland characteristics, and cropland characteristics.

[0030] A second determination module is configured to stack the remote sensing data set based on the target characteristics of the pixels to determine synthesized first image data.

[0031] A third determination module is configured to perform image segmentation on the first image data to determine segmented second image data.

[0032] An extraction module is configured to extract paddy field information of the to-be-identified region in a case where the target characteristics of the pixels of the second image data meet a preset condition.

[0033] The application also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor; when the processor executes the program, the method for extracting a paddy field based on a remote sensing cloud computing platform and phenological characteristics is realized.

[0034] The application also provides a non-transitory computer-readable storage medium having a computer program stored thereon; when the computer program is executed by a processor, the method for extracting a paddy field based on a remote sensing cloud computing platform and phenological characteristics is realized.

[0035] The application also provides a computer program product comprising a computer program which, when executed by a processor, implements the water field extraction method based on the remote sensing cloud computing platform and the phenological features.

[0036] The water field extraction method and device based on the remote sensing cloud computing platform and the phenological features provided by the application determine the water field features, wetland features and farmland features of the pixels of the remote sensing data set through the remote sensing data set of the crop in different growth periods of the to-be-identified region, stack and segment the remote sensing data, extract the water field information of the to-be-identified region according to the target features corresponding to the segmented image data, fully analyze the differences in the phenology of different land cover types, convert them into the feature parameters calculated by using the remote sensing data, and realize the accurate and effective extraction of the water field. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0038] Figure 1 is one of the flowcharts of the water field extraction method based on the remote sensing cloud computing platform and the phenological features provided by the embodiments of the application;

[0039] Figure 2 is the second flowchart of the water field extraction method based on the remote sensing cloud computing platform and the phenological features provided by the embodiments of the application;

[0040] Figure 3 is the structural schematic diagram of the water field extraction device based on the remote sensing cloud computing platform and the phenological features provided by the embodiments of the application;

[0041] Figure 4 is the structural schematic diagram of the electronic device provided by the embodiments of the application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the application more clear, the technical solutions in the application will be clearly and completely described below in combination with the drawings in the application. Obviously, the described embodiments are some of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0043] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second" are generally a class, not limited to the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / " generally indicates that the objects before and after are in an "or" relationship.

[0044] Figure 1 is one of the flowcharts of the paddy field extraction method based on the remote sensing cloud computing platform and the phenological characteristics provided by the embodiments of the present application, as shown in Figure 1 The paddy field extraction method based on the remote sensing cloud computing platform and the phenological characteristics provided by the embodiments of the present application comprises:

[0045] Step 101, determining the target features of the pixels of the remote sensing data set based on the remote sensing data set of the to-be-identified region in different growth periods of crops; the remote sensing data set comprises first remote sensing data in the transplanting period, second remote sensing data in the growth period and third remote sensing data in the harvesting period; the target features comprise paddy field features, wetland features and farmland features;

[0046] Step 102, stacking the remote sensing data set based on the target features of the pixels to determine the synthesized first image data;

[0047] Step 103, performing image segmentation on the first image data to determine the segmented second image data;

[0048] Step 104, extracting the paddy field information of the to-be-identified region in the case where the target features of the pixels of the second image data meet the preset conditions.

[0049] It should be noted that the execution subject of the paddy field extraction method based on the remote sensing cloud computing platform and the phenological characteristics provided by the present application can be an electronic device, a component in the electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. Illustratively, the mobile electronic device can be a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc., and the present application does not make specific limitations.

[0050] The paddy field extraction method based on the remote sensing cloud computing platform and the phenological characteristics provided by the present embodiment can be implemented on a remote sensing cloud computing platform, and the remote sensing cloud computing platform can be Google Earth Engine or PIE-Engine.

[0051] In step 101, target features of pixels of a remote sensing data set are determined based on remote sensing data sets of a to-be-identified region in different growth periods of crops.

[0052] The remote sensing data set includes first remote sensing data in a transplanting period, second remote sensing data in a growth period, and third remote sensing data in a harvesting period; and the target features include paddy field features, wetland features, and farmland features.

[0053] The different growth periods of crops can include a transplanting period, a growth period, and a harvesting period. Remote sensing data of the to-be-identified region in different growth periods of crops is collected to form a remote sensing data set.

[0054] The first remote sensing data in the transplanting period, the second remote sensing data in the growth period, and the third remote sensing data in the harvesting period of the to-be-identified region can be collected, and the first remote sensing data, the second remote sensing data, and the third remote sensing data are used to form a remote sensing data set.

[0055] Optionally, the remote sensing data is remote sensing data with a short revisit period or remote sensing data synthesized from multi-year remote sensing data. Generally, 10m-30m multi-spectral data with a revisit period less than 16 days is selected.

[0056] For example, all medium and high resolution remote sensing data of the transplanting period (DOY: 100-160), the growth period (DOY: 190-260) and the harvest period (DOY: 270-340) of the region to be identified can be acquired. The cloud cover of each image is less than 70%, and when the image of the transplanting period is insufficient, the remote sensing data of the past two years can be added.

[0057] In some embodiments, before determining the target features of the pixels of the remote sensing data set based on the remote sensing data sets of the region to be identified in different growth periods of crops, the method further comprises:

[0058] Performing cloud masking processing on the remote sensing data set.

[0059] After collecting the remote sensing data set, cloud masking processing can be performed on the remote sensing data set, for example, cloud detection algorithms or image quality control bands can be used to remove pixels blocked by clouds.

[0060] In some embodiments, determining the target features of the pixels of the remote sensing data set based on the remote sensing data sets of the region to be identified in different growth periods of crops comprises:

[0061] Determining a first flooded pixel corresponding to the transplanting period based on the bands of the first pixel corresponding to the first remote sensing data;

[0062] Determining the paddy field feature based on the flooding frequency of the first flooded pixel;

[0063] Determining a second flooded pixel corresponding to the harvest period based on the bands of the second pixel corresponding to the third remote sensing data;

[0064] Determining the wetland feature based on the flooding frequency of the second flooded pixel;

[0065] Determining the farmland feature based on the normalized difference vegetation index of the second remote sensing data and the normalized difference vegetation index of the third remote sensing data.

[0066] Optionally, during the transplanting of rice, the paddy field appears as water body information, and the shortwave infrared is sensitive to water information, and an increase in soil moisture will cause the overall downward movement of the band reflectance.

[0067] Therefore, whether a pixel of remote sensing data is a flooded pixel can be determined according to the bands corresponding to the pixel, that is, whether a pixel of remote sensing data is a flooded pixel can be determined by the following formula:

[0068]

[0069] Wherein, the flooded pixel represents the value of the pixel, the value of 1 represents the flooded pixel, and the value of 0 represents the non-flooded pixel, R represents the red band, G represents the green band, B represents the blue band, SWIR represents the short-wave infrared band, and NDVI represents the normalized vegetation index.

[0070] Optionally, the paddy field has a flooding signal at the transplanting period, so as to avoid the influence of abnormal values, the frequency of each first pixel being flooded by water at the transplanting period can be calculated as a paddy field feature.

[0071] Specifically, according to the first remote sensing data at the transplanting period, according to the bands of the first pixels corresponding to the first remote sensing data, whether the first pixel is a flooded pixel can be determined by using the above formula, that is, the first flooded pixel corresponding to the transplanting period can be determined.

[0072] According to the total number of the first pixels observed by the satellite at the transplanting period, the invalid observation number of the first pixels at the transplanting period due to cloud cover and other factors, and the number of the first flooded pixels, the flooding frequency of the first pixel can be determined.

[0073] The flooding frequency of the first pixel is determined as the paddy field feature.

[0074] In some embodiments, the expression of the paddy field feature is as follows:

[0075]

[0076] Wherein, the paddy index represents the paddy field feature, F (Transplanting) represents the flooding frequency of the first pixel, ∑Total represents the total number of the first pixels observed by the satellite at the transplanting period, ∑Bad represents the invalid observation number of the first pixels at the transplanting period, and ∑Flooded Pixel (Transplanting) represents the number of the first flooded pixels.

[0077] Optionally, during the transplanting period, wetlands or floating plants have similar flooding signals as the paddy field. During the harvesting period, the wetlands or floating plants also have the flooding signal, while the paddy field has no flooding signal because it is vegetation or crop residue. Therefore, the flooding frequency of the second pixel at the harvesting period is calculated to separate the wetlands and the floating plants.

[0078] Specifically, according to the third remote sensing data at the harvesting period, according to the bands of the second pixels corresponding to the third remote sensing data, whether the second pixel is a flooded pixel can be determined by using the above formula, that is, the second flooded pixel corresponding to the harvesting period can be determined.

[0079] And according to the total number of the second pixels observed by the satellite during the harvesting period, the number of invalid observations of the second pixels caused by cloud cover and other factors during the harvesting period, and the number of the second flooded pixels, the flooding frequency of the second pixels is determined.

[0080] And the flooding frequency of the second pixels is determined as the wetland feature.

[0081] In some embodiments, the expression of the wetland feature is as follows:

[0082]

[0083] Wherein, Wetland index represents the wetland feature, F(Harvesting) represents the flooding frequency of the second flooded pixels, ∑Total' represents the total number of the second pixels observed by the satellite during the harvesting period, ∑Bad' represents the number of invalid observations of the second pixels during the harvesting period, and ∑Flooded Pixel(Harvesting) represents the number of the second flooded pixels.

[0084] Optionally, considering that the phenological starting stage of farmland is generally later than that of natural vegetation, especially later than that of forest land. At the same time, due to artificial water and fertilizer input and management, farmland generally has a higher NDVI peak value.

[0085] In some embodiments, the expression of the farmland feature is as follows:

[0086]

[0087] Wherein, Cropland index represents the farmland feature, NDVI(Growing) represents the maximum value of the normalized vegetation index during the growing period, and NDVI(Transplanting) represents the maximum value of the normalized vegetation index during the transplanting period.

[0088] In order to avoid errors caused by outliers, the 80th percentile of the normalized vegetation index (NDVI) in each period can be taken as the maximum value.

[0089] In step 102, the remote sensing data set is stacked based on the target feature of the pixel to determine the synthesized first image data.

[0090] Through calculation, the target feature corresponding to each pixel can be obtained, including the paddy field feature, the wetland feature and the farmland feature.

[0091] According to the target feature of the pixel, the remote sensing data in the remote sensing data set is image stacked, that is, three single-band images are combined into a multi-band image, that is, the synthesized first image data is obtained.

[0092] In step 103, the first image data is image segmented to determine segmented second image data.

[0093] The classification or extraction of high-resolution remote sensing images often produces a "salt and pepper effect", and the use of a superpixel segmentation algorithm can effectively reduce the "salt and pepper effect".

[0094] In the embodiment of the application, a simple non-iterative clustering (SNIC) algorithm is used to segment the synthesized first image data to obtain the segmented second image data.

[0095] In step 104, if the target feature of the pixel of the second image data meets the preset condition, the paddy field information of the to-be-identified region is extracted.

[0096] The target feature of the pixel of the second image data is determined whether it meets the preset condition, and when the target feature of the pixel of the second image data meets the preset condition, it can be determined that the second image data is the image data corresponding to the paddy field, that is, the corresponding paddy field information can be extracted from the to-be-identified region.

[0097] The preset condition is as follows:

[0098]

[0099] Preferably, a1 is 0.1, a2 is 0.1, and a3 is 0.35. The values of a1, a2 and a3 can be set according to actual needs, and the application does not make specific limitations.

[0100] When the paddy field feature, the wetland feature and the farmland feature of the pixel of the second image data all meet the above preset condition, it can be determined that the second image data is the image data corresponding to the paddy field, that is, the corresponding paddy field information can be extracted from the to-be-identified region.

[0101] The paddy field extraction method based on the remote sensing cloud computing platform and the phenological feature provided in the embodiment of the application determines the paddy field feature, the wetland feature and the farmland feature of the pixel of the remote sensing data set through the remote sensing data set of the to-be-identified region in different growth periods of crops, and stacks and segments the remote sensing data, extracts the paddy field information of the to-be-identified region according to the target feature corresponding to the segmented image data, fully analyzes the differences in the phenology of different land cover types, converts them into feature parameters calculated using remote sensing data, and realizes accurate and effective extraction of paddy fields.

[0102] Figure 2is a flowchart of a second embodiment of a paddy field extraction method based on a remote sensing cloud computing platform and a phenological feature provided by the present application, as shown in Figure 2 The paddy field extraction method based on a remote sensing cloud computing platform and a phenological feature provided by the present application comprises the following steps.

[0103] Step 1, data acquisition

[0104] The present application is implemented on a remote sensing cloud computing platform, which can be Google EarthEngine or PIE-Engine.

[0105] The present application adopts remote sensing data with a short revisit period, or uses multi-year remote sensing data synthesis. Generally, 10m-30m multispectral data with a revisit period less than 16 days are selected.

[0106] The data preprocessing is as follows:

[0107] (1) Data screening.

[0108] All medium and high resolution remote sensing data of the working area in the transplanting period (DOY: 100-160), the vigorous growth period (DOY: 190-260) and the harvesting period (DOY: 270-340) are obtained. The cloud cover of each image is less than 70%, and when the image in the transplanting period is insufficient, the remote sensing data in the past two years can be added.

[0109] Specifically, the obtained data can include: first remote sensing data in the crop transplanting period, second remote sensing data in the growth period and third remote sensing data in the harvesting period.

[0110] (2) Cloud mask.

[0111] Cloud detection algorithms or image quality control bands are used to remove pixels obscured by clouds.

[0112] Step 2, detecting flooded pixels.

[0113] During the transplanting period of rice, the paddy field shows water body information, and the short-wave infrared is sensitive to water information. At the same time, the increase of soil humidity will cause the overall downward movement of the band reflectivity.

[0114] Therefore, whether a pixel is a flooded pixel can be determined according to the band corresponding to the pixel of the remote sensing data. The flooded pixel is a pixel submerged by water.

[0115] Step 3, extracting paddy field features.

[0116] The paddy field has a flooded signal in the transplanting period. In order to avoid the influence of abnormal values, the frequency of each pixel being flooded by water in the transplanting period is calculated as a paddy field feature.

[0117] Specifically, according to the first remote sensing data of the transplanting period, according to the band of the first pixel corresponding to the first remote sensing data, whether the first pixel is a flooded pixel can be detected, that is, the first flooded pixel corresponding to the transplanting period can be determined.

[0118] And according to the total number of the first pixel observed by the satellite during the transplanting period, the number of invalid observations of the first pixel caused by cloud cover and other factors during the transplanting period, and the number of the first flooded pixel, the flooding frequency of the first pixel is determined.

[0119] And the flooding frequency of the first pixel is determined as the paddy field feature.

[0120] Step 4, extract wetland features.

[0121] Because during the transplanting period, wetlands or floating plants and the like have similar flooding signals as paddy fields. And during the harvest period, wetlands or floating plants also have flooding signals, while paddy fields are vegetation or crop residues and do not have flooding signals, so the flooding frequency of the pixel during the harvest period is calculated to separate wetlands and floating plants.

[0122] Specifically, according to the third remote sensing data of the harvest period, according to the band of the second pixel corresponding to the third remote sensing data, whether the second pixel is a flooded pixel can be detected, that is, the second flooded pixel corresponding to the harvest period can be determined.

[0123] And according to the total number of the second pixel observed by the satellite during the harvest period, the number of invalid observations of the second pixel caused by cloud cover and other factors during the harvest period, and the number of the second flooded pixel, the flooding frequency of the second pixel is determined.

[0124] And the flooding frequency of the second pixel is determined as the wetland feature.

[0125] Step 5, extract farmland features.

[0126] Considering that the phenological beginning stage of farmland is generally later than natural vegetation, especially later than forest land. At the same time, due to artificial water and fertilizer input and management, farmland generally has a high NDVI peak value, so the farmland feature can be designed as:

[0127]

[0128] Wherein, NDVI(Growing) is the maximum value of NDVI in the growing period, and NDVI(Transplanting) is the maximum value of NDVI in the transplanting period. In order to avoid errors caused by outliers, it is recommended to take the 80% percentile of NDVI in each period as the maximum value.

[0129] Step 6, image stacking.

[0130] The calculated paddy field feature, wetland feature and farmland feature are stacked into an image, and the three single-band images are combined into a multi-band image.

[0131] Step 7, image segmentation.

[0132] The high-resolution remote sensing image is classified or extracted, and the“salt and pepper effect” is effectively reduced by using a superpixel segmentation algorithm. The synthesized image is segmented by using a simple non-iterative clustering (SNIC) algorithm.

[0133] Step 8, extracting a paddy field by using a decision rule.

[0134] The segmented image is subjected to a decision rule to realize paddy field extraction, and the decision rule is as follows:

[0135] (1) Paddy field feature: Paddy index > a1

[0136] (2) Wetland feature: Wetland index < a2

[0137] (3) Farmland feature: Cropland index > a3

[0138] Wherein, the default value of a1 is 0.1, the default value of a2 is 0.1, and the default value of a3 is 0.35.

[0139] The paddy field extraction method based on the remote sensing cloud computing platform and the phenological feature provided in the embodiment of the present application determines the paddy field feature, the wetland feature and the farmland feature of the pixel of the remote sensing data set through the remote sensing data set of the crop in different growth periods, stacks and segments the remote sensing data, extracts the paddy field information of the to-be-identified region according to the target feature corresponding to the segmented image data, fully analyzes the differences of different land cover types in the phenology, converts the differences into the feature parameters calculated by using the remote sensing data, and realizes the accurate and effective extraction of the paddy field.

[0140] The paddy field extraction device based on the remote sensing cloud computing platform and the phenological feature provided in the embodiment of the present application is described below. The paddy field extraction device based on the remote sensing cloud computing platform and the phenological feature described below can be correspondingly referred to the paddy field extraction method based on the remote sensing cloud computing platform and the phenological feature described above.

[0141] Figure 3 The paddy field extraction device based on the remote sensing cloud computing platform and the phenological feature provided in the embodiment of the present application is described below. The paddy field extraction device based on the remote sensing cloud computing platform and the phenological feature described below can be correspondingly referred to the paddy field extraction method based on the remote sensing cloud computing platform and the phenological feature described above. Figure 3 As shown in FIG. 1, the paddy field extraction device based on the remote sensing cloud computing platform and the phenological feature provided in the embodiment of the present application comprises:

[0142] The first determining module 310 is configured to determine a target feature of a pixel of a remote sensing data set based on a to-be-identified region in different growth periods of crops, wherein the remote sensing data set comprises first remote sensing data in a transplanting period, second remote sensing data in a growth period and third remote sensing data in a harvesting period, and the target feature comprises a paddy field feature, a wetland feature and a farmland feature.

[0143] The second determining module 320 is configured to stack the remote sensing data set based on the target feature of the pixel to determine synthesized first image data.

[0144] The third determining module 330 is configured to perform image segmentation on the first image data to determine segmented second image data.

[0145] The extracting module 340 is configured to extract paddy field information of the to-be-identified region in a case where a target feature of a pixel of the second image data meets a preset condition.

[0146] It should be noted that the paddy field extraction device based on the remote sensing cloud computing platform and the phenological feature provided in the embodiment of the present application can realize all the method steps achieved by the paddy field extraction method based on the remote sensing cloud computing platform and the phenological feature, and can achieve the same technical effects. Therefore, the same parts and beneficial effects of the method embodiment in the embodiment will not be described in detail.

[0147] Optionally, the first determining module 310 is specifically configured to:

[0148] determine a first waterlogged pixel corresponding to the transplanting period based on a band of a first pixel corresponding to the first remote sensing data;

[0149] determine the paddy field feature based on a quantity of the first waterlogged pixel;

[0150] determine a second waterlogged pixel corresponding to the harvesting period based on a band of a second pixel corresponding to the third remote sensing data;

[0151] determine the wetland feature based on a quantity of the second waterlogged pixel;

[0152] determine the farmland feature based on a normalized vegetation index of the second remote sensing data and a normalized vegetation index of the third remote sensing data.

[0153] Optionally, the device further comprises a processing module configured to:

[0154] perform cloud mask processing on the remote sensing data set.

[0155] Optionally, the paddy field feature is expressed as follows:

[0156]

[0157] wherein Paddy index represents the paddy field feature, F(Transplanting) represents the flooding frequency of the first pixel, ∑Total represents the total number of observations of the first pixel by satellite during the transplanting period, ∑Bad represents the number of invalid observations of the first pixel during the transplanting period, and ∑Flooded Pixel(Transplanting) represents the number of second flooded pixels.

[0158] Alternatively, the expression of the wetland feature is as follows:

[0159]

[0160] wherein Wetland index represents the wetland feature, F(Harvesting) represents the flooding frequency of the second pixel, ∑Total' represents the total number of observations of the second pixel by satellite during the harvesting period, ∑Bad' represents the number of invalid observations of the second pixel during the harvesting period, and ∑Flooded Pixel(Harvesting) represents the number of second flooded pixels.

[0161] Alternatively, the expression of the cropland feature is as follows:

[0162]

[0163] wherein Cropland index represents the cropland feature, NDVI(Growing) represents the maximum value of the normalized vegetation index during the growing period, and NDVI(Transplanting) represents the maximum value of the normalized vegetation index during the transplanting period.

[0164] Figure 4 An example of a schematic diagram of the physical structure of an electronic device is shown in FIG. 1. Figure 4As shown, the electronic device can include a processor 410, a communications interface 420, a memory 430, and a communications bus 440, wherein the processor 410, the communications interface 420, and the memory 430 complete mutual communication through the communications bus 440. The processor 410 can invoke a logical instruction in the memory 430 to execute a paddy field extraction method based on a remote sensing cloud computing platform and a phenological feature, the method including: determining a target feature of a pixel of a remote sensing data set of a to-be-identified region in different growth periods of crops based on the remote sensing data set; the remote sensing data set includes first remote sensing data in a transplanting period, second remote sensing data in a growth period, and third remote sensing data in a harvesting period; the target feature includes a paddy field feature, a wetland feature, and a farmland feature; stacking the remote sensing data set based on the target feature of the pixel to determine a synthesized first image data; performing image segmentation on the first image data to determine a segmented second image data; and in a case where the target feature of the pixel of the second image data meets a preset condition, extracting paddy field information of the to-be-identified region.

[0165] In addition, the logical instructions in the memory 430 described above can be implemented in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0166] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program being stored on a non-transitory computer readable storage medium, and the computer program being executable by a processor to enable a computer to perform the paddy field extraction method based on a remote sensing cloud computing platform and a phenological feature, the method comprising: determining target features of pixels of a remote sensing data set of a to-be-identified region in different growth periods of crops based on the remote sensing data set, wherein the remote sensing data set comprises first remote sensing data in a transplanting period, second remote sensing data in a growth period, and third remote sensing data in a harvesting period; the target features comprise paddy field features, wetland features, and farmland features; stacking the remote sensing data set based on the target features of the pixels to determine synthesized first image data; performing image segmentation on the first image data to determine segmented second image data; and extracting paddy field information of the to-be-identified region in a case where the target features of the pixels of the second image data meet preset conditions.

[0167] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having a computer program stored thereon, the computer program being executable by a processor to implement a paddy field extraction method based on a remote sensing cloud computing platform and a phenological feature, the method comprising: determining target features of pixels of a remote sensing data set of a to-be-identified region in different growth periods of crops based on the remote sensing data set, wherein the remote sensing data set comprises first remote sensing data in a transplanting period, second remote sensing data in a growth period, and third remote sensing data in a harvesting period; the target features comprise paddy field features, wetland features, and farmland features; stacking the remote sensing data set based on the target features of the pixels to determine synthesized first image data; performing image segmentation on the first image data to determine segmented second image data; and extracting paddy field information of the to-be-identified region in a case where the target features of the pixels of the second image data meet preset conditions.

[0168] The device embodiments described above are merely illustrative, wherein the units illustrated as separate components can or can not be physically separate, and the components illustrated as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0169] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0170] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for extracting paddy fields based on a remote sensing cloud computing platform and phenological features, characterized in that, The method comprises the following steps: determining target features of pixels of a remote sensing data set based on remote sensing data sets of the to-be-identified region in different growth periods of crops; the remote sensing data set comprises first remote sensing data of a transplanting period, second remote sensing data of a growth period, and third remote sensing data of a harvest period; and the target features comprise paddy field features, wetland features, and cropland features; stacking the remote sensing data set based on the target features of the pixels to determine first image data; performing image segmentation on the first image data to determine second image data after segmentation; extracting paddy field information of the to-be-identified region in a case where target features of pixels of the second image data meet a preset condition; the step of determining target features of pixels of a remote sensing data set based on remote sensing data sets of the to-be-identified region in different growth periods of crops comprises the following steps: determining first flooded pixels corresponding to the transplanting period based on bands of first pixels corresponding to the first remote sensing data; determining the paddy field features based on a number of the first flooded pixels; determining second flooded pixels corresponding to the harvest period based on bands of second pixels corresponding to the third remote sensing data; determining the wetland features based on a number of the second flooded pixels; determining the cropland features based on a normalized vegetation index of the second remote sensing data and a normalized vegetation index of the third remote sensing data. 2.The paddy field extraction method based on remote sensing cloud computing platform and phenological features according to claim 1, characterized in that, Before the step of determining target features of pixels of a remote sensing data set based on remote sensing data sets of the to-be-identified region in different growth periods of crops, the method further comprises the following step: performing cloud mask processing on the remote sensing data set. 3.The paddy field extraction method based on remote sensing cloud computing platform and phenological features according to claim 1, characterized in that, An expression of the paddy field features is as follows: wherein Paddy index represents the paddy field features, F(Transplanting) represents a flooded frequency of the first pixels, ∑Total represents a total number of observations of the first pixels by a satellite in the transplanting period, ∑Bad represents a number of invalid observations of the first pixels in the transplanting period, and ∑Flooded Pixel(Transplanting) represents the number of the first flooded pixels. 4.The paddy field extraction method based on remote sensing cloud computing platform and phenological features according to claim 1, characterized in that, An expression of the wetland features is as follows: where Wetland index represents the wetland characteristic, F(Harvesting) represents the frequency of flooding of the second pixel, ∑Total ′ represents the total number of observations of the second pixel by the satellite at the harvest period, ∑Bad ′ represents the number of invalid observations of the second pixel at the harvest period, and ∑Flooded Pixel(Harvesting) represents the number of flooded pixels at the harvest period. 5.The paddy field extraction method based on remote sensing cloud computing platform and phenological features according to claim 1, characterized in that, An expression of the cropland features is as follows: wherein Cropland index represents the cropland features, NDVI(Growing) represents a maximum value of a normalized vegetation index in the growth period, and NDVI(Transplanting) represents a maximum value of a normalized vegetation index in the transplanting period.

6. A paddy field extraction device based on a remote sensing cloud computing platform and a phenological feature, characterized by, The method comprises the following steps: a first determining module is configured to determine target features of pixels of a remote sensing data set based on remote sensing data sets of the to-be-identified region in different growth periods of crops; the remote sensing data set comprises first remote sensing data of a transplanting period, second remote sensing data of a growth period, and third remote sensing data of a harvest period; and the target features comprise paddy field features, wetland features, and cropland features; a second determining module is configured to stack the remote sensing data set based on the target features of the pixels to determine first image data; a third determining module is configured to perform image segmentation on the first image data to determine second image data after segmentation; and a fourth determining module is configured to extract paddy field information of the to-be-identified region in a case where target features of pixels of the second image data meet a preset condition. The extraction module is configured to extract paddy field information of the to-be-identified region when it is determined that a target feature of a pixel of the second image data meets a preset condition. The target feature of the pixel of the remote sensing data set is determined based on the remote sensing data set of the to-be-identified region in different growth periods of crops, and includes: Determine the first flooded pixel corresponding to the transplanting period based on the band of the first pixel corresponding to the first remote sensing data; Determine the paddy field feature based on the number of the first flooded pixel; Determine the second flooded pixel corresponding to the harvesting period based on the band of the second pixel corresponding to the third remote sensing data; Determine the wetland feature based on the number of the second flooded pixel; Determine the farmland feature based on the normalized difference vegetation index of the second remote sensing data and the normalized difference vegetation index of the third remote sensing data.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, The processor executes the program to realize the paddy field extraction method based on the remote sensing cloud computing platform and the phenology feature according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the paddy field extraction method based on the remote sensing cloud computing platform and the phenology feature according to any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the paddy field extraction method based on the remote sensing cloud computing platform and the phenology feature according to any one of claims 1 to 5.

Images