A method and system for identifying the heading stage of crops based on radar time series observations

Abstract

The present application belongs to the technical field of spaceborne radar data processing, and particularly relates to a crop heading stage identification method and system based on radar time series observation, which obtains a depolarization index through index calculation, realizes periodic parameter extraction of the time series depolarization index through harmonic decomposition, determines the dominant periodic component of the time series depolarization index through weight analysis, identifies the crop heading stage through wave peak detection, and realizes crop heading stage mapping in combination with land use products. The present application determines the dominant periodic component of the time series depolarization index through weight analysis, identifies the crop heading stage through wave peak detection, and realizes crop heading stage mapping in combination with land use products, and is suitable for large-area crop heading stage identification in the case of complex and diverse farming practices and frequent cloud and fog meteorological interference, and solves the three major problems of lack of prior phenological information, heading stage feature ambiguity and non-heading stage feature interference in the current crop heading stage identification using radar data.

Description

Technical Field

[0001] This invention belongs to the field of spaceborne radar data processing technology, and in particular relates to a method and system for identifying the heading stage of crops based on radar time-series observation. Background Technology

[0002] Heading refers to the physiological phenomenon of a crop plant developing a spike or inflorescence, a crucial step in the crop growth cycle that marks the beginning of pollination and seed formation. The agricultural sector typically monitors the heading stage of crops closely to determine the optimal harvest time for maximum yield. The heading stage is usually related to crop variety, climatic conditions, and cultivation practices. Different crops can be planted on arable land in different seasons to maximize land productivity. In such cases, arable land may have multiple crop rotations within a year, resulting in multiple heading stages. Satellite remote sensing is an effective means of identifying crop heading stages; identifying the heading stages of crops in different rotations requires year-round remote sensing data. Optical satellites are traditional remote sensing platforms capable of accurately detecting changes in crop growth status. However, optical satellites are highly susceptible to cloud cover and fog, making year-round remote sensing observations difficult. Radar satellites, on the other hand, can provide stable, high-quality imaging unaffected by weather conditions, making them an important alternative data source for identifying crop heading stages.

[0003] Radar time-series observations can acquire temporal scattering data of cultivated land, analyze changes in crop plant and soil traits, and establish a relationship between the characteristics of time-series scattering curves and the crop heading stage. The time-series feature discrimination method is the most commonly used approach. Based on prior phenological information of crops, it conducts radar time-series observations during the crop growing season, constructs a time-series scattering curve for the crop, and uses multi-order differentiation and local trend analysis to detect curve characteristics, thereby identifying the crop heading stage. However, agricultural planting areas have complex and diverse crop planting structures and vastly different field management practices, making it extremely difficult to obtain prior phenological information. Secondly, the characteristics of the time-series scattering curve at the crop heading stage are ambiguous. It is generally believed that the peak of the time-series scattering curve corresponds to the crop heading stage, but when the crop canopy density is high, the peak characteristics will shift over time. Finally, scattering coherence noise, weed growth, and human activities can sometimes cause spurious peaks in the time-series scattering curve; these non-heading stage characteristics can interfere with the identification of the crop heading stage. These three issues severely restrict the operational application of radar time-series observation in identifying the heading stage of crops.

[0004] Based on the above analysis, the problems and defects of the existing technology are as follows: the existing methods lack prior phenological information, the characteristics of the heading period are vague, and the interference of non-heading period characteristics remains to be solved. Summary of the Invention

[0005] In view of the problems in the prior art, the present application provides a crop heading stage identification method and system based on radar time series observation.

[0006] The present application is implemented in a crop heading stage identification method based on radar time series observation, a depolarization index is obtained through index calculation, periodicity parameter extraction of the time series depolarization index is realized through harmonic decomposition, the dominant periodic component of the time series depolarization index is determined through weight analysis, the crop heading stage is identified through wave peak detection, and crop heading stage mapping is realized in combination with land use products.

[0007] Further, the crop heading stage identification method based on radar time series observation includes the following specific steps:

[0008] S1, index calculation:

[0009] The VH (Vertical-Horizontal) and VV (Vertical-Vertical) polarization data of annual radar time series observation are obtained, and the depolarization index (DI, Depolarization Index) is calculated using the following formula:

[0010] DI = 10 VH / 10 / (10 VH / 10 + 10 VV / 10 )

[0011] S2, harmonic decomposition:

[0012] For the time series depolarization index D[t], t is the normalized annual accumulated day, and harmonic decomposition is performed using the following formula:

[0013]

[0014] a represents the time series mean, A i represents the amplitude of the i-th cosine term, represents the phase of the i-th cosine term. The values of a, A i and are obtained using least squares fitting;

[0015] S3, weight analysis:

[0016] The weight W i of the i-th cosine term is defined as follows:

[0017]

[0018] The maximum weight W Max of the three cosine terms is:

[0019] W Max = max i Wi

[0020] When the weight W of the i-th cosine term i Equal to the maximum weight W of the three cosine terms Max At that time, that is:

[0021] W i =W Max

[0022] The dominant periodic component is the i-th cosine term;

[0023] S4, Peak Detection:

[0024] The first-order cosine term has one peak, and the peak date C1 is calculated by the following formula:

[0025]

[0026] The second-order cosine term has two peaks, with peak date C. 2a and C 2b Calculated by the following formula:

[0027]

[0028] The third-order cosine term has three peaks, with peak date C. 3a C 3b and C 3c Calculated by the following formula:

[0029]

[0030] S5, Crop heading stage mapping:

[0031] The system acquires land use products and extracts cultivated land areas. It then determines the dominant periodic components pixel by pixel within the cultivated land areas, thereby determining the number and date of peaks. Finally, it determines whether the cultivated land is in the heading stage of crops on a monthly basis, thus enabling crop heading stage mapping.

[0032] Furthermore, in S3, harmonic decomposition treats the time series DI as a synthesis of a constant component and three-order cosine components. Thus, each peak in the time series DI is a synthesis of the peaks of the three-order cosine components. The local trend of the peak can be approximated as the slope from the trough to the peak. This slope can be represented in the cosine component as the ratio of amplitude to order i, representing the weight of the cosine component. The cosine component with the largest weight describes the dominant periodicity of the time series DI.

[0033] Furthermore, in S4, the peak position of the cosine component depends on the phase and order i, and the center point of the cosine component is... The date corresponds to the peak of the second-order cosine component and the troughs of the first and third-order cosine components. Using the center date as the initial point and the intervals as integer multiples of half the period of the cosine component, the dates of the remaining peaks can be obtained through step operations.

[0034] Furthermore, in S5, the premise for identifying the crop heading period is to know the specific cultivated land area. By performing index calculation, harmonic decomposition, weight analysis, and peak detection on a pixel-by-pixel basis, the crop heading period of the cultivated land area is obtained. In order to better display the temporal distribution characteristics of the crop heading period, a monthly mapping strategy is adopted to identify whether the cultivated land pixel is in the crop heading period.

[0035] Another objective of this invention is to provide a crop heading date identification system based on radar time-series observation, which applies the aforementioned method for identifying crop heading dates based on radar time-series observation, comprising:

[0036] Index calculation module: used to acquire VH (Vertical-Horizontal) and VV (Vertical-Vertical) polarization data from annual radar time-series observations and to calculate the depolarization index (DI).

[0037] Harmonic decomposition module: used to perform harmonic decomposition on the timing depolarization index;

[0038] Weight Analysis Module: Used for performing weight analysis;

[0039] Peak detection module: used for peak detection of the third-order cosine term;

[0040] The crop heading period mapping module is used to obtain land use products and extract cultivated land areas; determine the dominant periodic components pixel by pixel in the cultivated land area, and then determine the number and date of peaks; determine whether the cultivated land is in the crop heading period month by month, and realize crop heading period mapping.

[0041] Another object of the present invention is to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, and when the computer program is executed by the processor, causing the processor to perform the steps of the crop heading period identification method based on radar time-series observation.

[0042] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the crop heading stage identification method based on radar time-series observation.

[0043] Another object of the present application is to provide an information data processing terminal for implementing the crop heading date identification system based on radar time series observation.

[0044] In combination with the above technical solutions and solved technical problems, the technical solutions of the present application have the following advantages and positive effects:

[0045] Firstly, the present application successfully solves the important technical problems in the current crop heading date identification using radar time series observation. In view of the lack of prior phenological information, the present application analyzes the radar time series observation of the whole year, rather than only the radar time series observation of the crop growing season. In view of the feature ambiguity problem of the heading date, the present application introduces the depolarization index and harmonic decomposition technology, so that the stable crop heading date is expressed as the peak of the time series curve. In view of the feature interference problem of the non-heading date, the present application introduces the weight analysis and peak detection technology, which eliminates the influence of non-heading date features by determining the dominant periodicity of the time series curve, and quickly corresponds the peak of the dominant periodicity component to the crop heading date. In summary, the present application successfully solves the three core problems in the prior art, making the identification of crop heading date more efficient and accurate.

[0046] The present application does not need to obtain prior phenological information of the research area in advance, does not need to perform window filtering on the radar time series observation, the time series curve peak feature of the crop heading date is significant, does not need to perform local trend analysis and false peak discrimination on the time series curve, and the peak date of the crop heading date can be quickly obtained through phase operation.

[0047] Secondly, the expected income and commercial value of the technical solutions of the present application after transformation are as follows: the present application can be used by commercial consulting companies to provide high-value services such as farmland management consulting, intelligent agricultural monitoring and operation strategy analysis; can help agricultural operators to reasonably arrange farmland management activities according to the heading date information, reduce the cost of fertilization, irrigation, and pest control, and improve the yield and quality of crops; can help the food supply chain to more efficiently arrange the harvesting and distribution of agricultural products, reduce resource waste and reduce environmental impact; can help the insurance department to more accurately assess and manage agricultural risks, and timely develop agricultural insurance policies; can help the agricultural management department to more comprehensively estimate the grain yield and supply time, prevent abnormal fluctuations in the market price of agricultural products, and ensure food security.

[0048] The technical solution of the present application fills the technical gap in the industry at home and abroad: the present application proposes a scheme for identifying the heading stage of crops using radar time series observation data throughout the year, solving the problem of the difficulty of expanding the research area due to the lack of prior phenological information. The present application proposes a new depolarization index to more accurately track crop growth, solving the problem of blurred features of the heading stage of crops caused by the use of only single polarization data. The present application proposes a new time series data analysis approach, including steps such as harmonic decomposition, weight analysis and wave peak detection, solving the problem of non-heading stage feature interference caused by scattering coherent noise, weed growth and human activities.

[0049] Thirdly, the crop heading stage identification method provided by the present application represents a major progress in the application of remote sensing technology in precision agriculture. Through the analysis and processing of radar time series data, this method can accurately identify the heading stage, thereby effectively monitoring the growth and development of crops. The following summarizes the significant technical progress brought about by this method:

[0050] 1. High precision in time identification:

[0051] The depolarization index calculated using VH and VV polarization data provides accurate time series data for monitoring crop growth.

[0052] The application of harmonic decomposition extracts the periodic changes in the time series depolarization index, which is crucial for capturing and understanding the crop growth cycle.

[0053] 2. High efficiency of data analysis:

[0054] Least squares fitting is an effective method in mathematics for processing observation data to find the best function match, which can accurately estimate the periodic parameters.

[0055] The introduction of weight analysis can identify the main periodic components that affect the identification of the heading stage, improving the reliability of the identification method.

[0056] 3. Accurate detection of the heading stage:

[0057] Wave peak detection is an accurate means of determining the key stages of crop growth, which helps to determine the heading stage and plays an important role in crop growth monitoring and prediction.

[0058] The use of multiple order cosine terms enables the detection of multiple growth peaks occurring throughout the year, adapting to different crops and diverse growth conditions.

[0059] 4. Integration of land use data:

[0060] Combining land use products can more accurately distinguish between cultivated land and non-cultivated land, and only identify the heading stage in relevant areas, improving the efficiency and accuracy of the analysis.

[0061] Mapping of heading dates at the plot level is crucial for understanding regional agricultural production patterns and optimizing resource allocation.

[0062] Practical value:

[0063] Agricultural management:

[0064] Timely understanding of crop growth conditions provides scientific basis for agricultural production management, such as irrigation, fertilization, pest control, etc.

[0065] Harvest prediction:

[0066] Through the identification of heading date, the crop harvest time and yield can be estimated, which has important influence on food security and market supply.

[0067] Agricultural insurance:

[0068] Provide accurate crop growth data for agricultural insurance, reduce the risk of insurance companies when paying claims.

[0069] Agricultural policy making:

[0070] Government can better plan food reserves and agricultural support policies according to the timing data of crop growth.

[0071] Climate change research:

[0072] Long-term observation of crop growth data can be used to study the impact of climate change on agriculture.

[0073] In summary, the method of identifying crop heading date based on radar timing observation improves the accuracy of crop growth monitoring and provides an efficient tool for agricultural production management, which is of great significance for improving the sustainability of agricultural production, increasing food yield and quality, and reducing environmental impact. BRIEF DESCRIPTION OF DRAWINGS

[0074] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0075] Figure 1 is the flow chart of the method for identifying crop heading date based on radar timing observation provided by the embodiments of the present application;

[0076] Figure 2 is the structural diagram of the system for identifying crop heading date based on radar timing observation provided by the embodiments of the present application;

[0077] Figure 3The timing depolarization index graph is provided by an embodiment of the present application.

[0078] Figure 4 The harmonic decomposition graph is provided by an embodiment of the present application.

[0079] Figure 5 The weight analysis and wave peak detection graph is provided by an embodiment of the present application.

[0080] Figure 6 The crop heading stage identification graph is provided by an embodiment of the present application. DETAILED DESCRIPTION

[0081] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application is further described in detail below in combination with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0082] In order to solve the problems in the prior art, the present application provides a crop heading stage identification method and system based on radar timing observation, which is described in detail below in combination with the drawings.

[0083] Embodiment 1: precise rice heading stage monitoring

[0084] 1. Data acquisition:

[0085] Synthetic aperture radar (SAR) is used to obtain VH and VV polarization timing data of the rice planting area throughout the year from different satellite platforms.

[0086] 2. Depolarization index calculation:

[0087] The original radar data is processed by using the DI calculation formula to obtain a depolarization index sequence about the growth state of crops.

[0088] 3. Harmonic decomposition:

[0089] The depolarization index timing data is subjected to harmonic decomposition to capture the rice growth cycle.

[0090] 4. Weight analysis and periodic component identification:

[0091] Weight analysis is performed to determine which order cosine term best represents the growth cycle of rice.

[0092] 5. Wave peak detection and heading stage prediction:

[0093] According to the dominant periodic component, the wave peak detection algorithm is used to predict the heading date of rice.

[0094] 6. Result application and verification:

[0095] The land use data is combined to divide the cultivated areas, and then the wave peak date is applied within these areas to identify the heading stage.

[0096] The heading stage prediction results obtained from radar time series observations are verified using ground survey data.

[0097] Embodiment 2: Wheat crop growth cycle monitoring

[0098] 1. Data collection:

[0099] VH and VV polarization data of the wheat planting area are obtained through multiple time point radar satellites, such as Sentinel-1.

[0100] 2. Depolarization index extraction:

[0101] Using radar data, the depolarization index sequence is calculated according to the formula.

[0102] 3. Harmonic decomposition and cycle analysis:

[0103] Harmonic decomposition is performed on the depolarization index to extract the periodic changes consistent with the wheat growth cycle.

[0104] 4. Determination of important periodic components:

[0105] By analyzing the weights of different period terms, determine which one best reflects the key period of wheat growth.

[0106] 5. Wave peak identification of heading stage:

[0107] According to the determined periodic component, wave peak detection technology is used to predict the heading stage of wheat.

[0108] 6. Mapping and decision support:

[0109] Land use data is used for mapping to determine the spatial distribution of the heading stage.

[0110] The results are used to guide agricultural activities such as irrigation, fertilization planning, etc. to optimize the production process.

[0111] The specific implementation schemes in these two embodiments provide a complete process from data collection to result application, which can accurately monitor the growth cycle of different crops and provide strong support for agricultural production and management decisions. The crop heading stage identification method based on radar time series observations provided in the embodiments of the invention obtains the depolarization index through index calculation, extracts the periodic parameters of the time series depolarization index through harmonic decomposition, determines the dominant periodic component of the time series depolarization index through weight analysis, identifies the heading stage of crops through wave peak detection, and realizes the mapping of the heading stage of crops in combination with land use products.

[0112] AsFigure 1 The crop heading date identification method based on radar time series observation includes the following specific steps:

[0113] S1, index calculation:

[0114] The VH (Vertical-Horizontal) and VV (Vertical-Vertical) polarization data of the annual radar time series observation are obtained, and the depolarization index (DI) is calculated using the following formula:

[0115] DI = 10 VH / 10 / (10 VH / 10 + 10 VV / 10 )

[0116] S2, harmonic decomposition:

[0117] For the time series depolarization index D[t], t is the normalized annual accumulated day, and the harmonic decomposition is performed using the following formula:

[0118]

[0119] a represents the time series mean, A i represents the amplitude of the i-th cosine term, represents the phase of the i-th cosine term. The values of a, A i and are obtained by least squares fitting;

[0120] S3, weight analysis:

[0121] The weight W i of the i-th cosine term is defined as follows:

[0122]

[0123] The maximum weight W Max of the three cosine terms is:

[0124] W Max = max i W i

[0125] When the weight W i of the i-th cosine term is equal to the maximum weight W Max of the three cosine terms, that is:

[0126] W i = W Max

[0127] The dominant periodic component is the i-th cosine term;

[0128] S4, peak detection:

[0129] The 1st order cosine term has one peak, and the peak date C1 is calculated by:

[0130]

[0131] The 2nd order cosine term has two peaks, and the peak dates C 2a and C 2b are calculated by:

[0132]

[0133] The 3rd order cosine term has three peaks, and the peak dates C 3a , C 3b and C 3c are calculated by:

[0134]

[0135] S5, mapping of crop heading period:

[0136] Obtain the land use product, extract the cultivated land region; determine the dominant periodic component in the cultivated land region pixel by pixel, and then determine the number and date of the peaks; judge whether the cultivated land is in the crop heading period month by month to realize the mapping of the crop heading period.

[0137] Index calculation principle: VH polarization represents the vibration of electromagnetic waves in the vertical and horizontal directions, and VV polarization represents the vibration of electromagnetic waves in the vertical direction. VH polarization is sensitive to the directionality and roughness of the target, and VV polarization is more sensitive to the vertical characteristics of the ground object. The sum of VH and VV represents the total echo energy of the radar, and the depolarization index DI is the ratio of VH to the total echo energy. DI represents the proportion of the vertical transmitted signal twisted into horizontal vibration signal by the ground object, which is the highest in the crop heading period, so DI presents a peak in this period.

[0138] Harmonic decomposition principle: cultivated land can have multiple crop planting rounds in a year. For each crop planting round, DI will go through an approximate cosine process from low (germination) to high (heading) and then low (harvesting). This process can be described by the cosine component in harmonic decomposition. Generally speaking, it takes more than 100 days for crops to grow from germination to harvesting, and there can be up to three rounds of crop planting in a year, so three orders of cosine components are used in harmonic decomposition. Cosine components can only represent the fluctuation pattern of DI, not the magnitude level, so a constant component is used to represent the annual average of DI.

[0139] Weight analysis principle: the harmonic decomposition regards the time series DI as the combination of a constant component and three cosine components of different orders, then each peak in the time series DI is the combination of the peaks of the three cosine components. The local trend of the peak can be approximately expressed as the slope from the trough to the peak, which can be represented as the ratio of the amplitude and the order i in the cosine component, representing the weight of the cosine component. The cosine component with the largest weight describes the dominant periodicity of the time series DI.

[0140] Peak detection principle: the peak position of the cosine component depends on the phase and the order i, and the central point date of the cosine component is This date corresponds to the peak of the 2nd order cosine component, and the troughs of the 1st and 3rd order cosine components. Taking the central point date as the initial point and the integer multiple of the half period of the cosine component as the interval, the dates of the remaining peaks can be obtained by stepping operation.

[0141] Crop heading date mapping principle: the prerequisite for crop heading date identification is to know the certain farmland area, and there are a large number of land use products that can provide spatial distribution information of farmland. Through pixel-by-pixel index calculation, harmonic decomposition, weight analysis and peak detection, the crop heading date of the farmland area can be obtained. In order to better show the time distribution characteristics of the crop heading date, the pixel-by-pixel mapping strategy is adopted to identify whether the farmland pixel is in the crop heading date.

[0142] As shown in Figure 2 The crop heading date identification system based on radar time series observation provided by the embodiment of the present application comprises:

[0143] Index calculation module: used for obtaining the VH(Vertical-Horizontal) and VV(Vertical-Vertical) polarization data of the annual radar time series observation, and performing DI(Depolarization Index) calculation;

[0144] Harmonic decomposition module: used for performing harmonic decomposition on the time series DI;

[0145] Weight analysis module: used for weight analysis;

[0146] Peak detection module: used for peak detection of the 3rd order cosine term;

[0147] Crop heading date mapping module: used for obtaining the land use product, extracting the farmland area, determining the dominant periodic component pixel by pixel in the farmland area, and then determining the number and date of the peaks; judging whether the farmland is in the crop heading date month by month to realize the mapping of the crop heading date.

[0148] A method for identifying crop heading date based on radar time series observation and harmonic analysis, the development environment of the embodiment is GEE (Google earth engine), and the programming language is JavaScript.

[0149] Step 1, retrieve Sentinel-1 data of Changxing County, Huzhou City in 2021, and use the image.pow and image.divide functions to construct the time series of the depolarization index.

[0150] Step 2, use the image.cos and image.linearRegression functions to obtain the undetermined coefficients a, Ai and i of each cosine component, and realize the harmonic decomposition of the time series of the depolarization index.

[0151] Step 3, use the image.multiply and reduce(‘sum’) functions to obtain the weight of each cosine component, use the reduce(‘max’) and image.eq functions to determine the dominant periodic component, and use the image.abs and image.add functions to perform peak detection and peak date calculation.

[0152] Step 4, retrieve the ESA WorldCover2021 land use product, and use the image.mask function to extract the cultivated land area. Use the image.gte and image.lte functions to determine the dominant periodic component and calculate the peak date pixel by pixel in the cultivated land area, and use the image.and and image.or functions to realize the identification of crop heading date.

[0153] In the embodiment, the time series Sentinel-1 radar data and ESA WorldCover2021 data are processed, Figure 3 the time series depolarization index chart of the embodiment, Figure 4 the harmonic decomposition chart of the embodiment, Figure 5 the weight analysis and peak detection chart of the embodiment, Figure 6 the crop heading date identification chart of the embodiment.

[0154] The problems and defects of the prior art are that the prior phenology information is lacking, the heading stage feature is fuzzy, and the non-heading stage feature is interfered. Firstly, the crop planting structure in the agricultural planting area is complex and diverse, the field management scheme is greatly different, and the prior phenology information of the crops is extremely difficult to obtain. Secondly, the feature of the time series scattering curve in the heading stage of the crops is also fuzzy, and it is generally considered that the peak of the time series scattering curve corresponds to the heading stage of the crops, but when the canopy density of the crops is high, the peak feature is time-shifted. Finally, the scattering coherent noise, weed growth and human activities sometimes cause the appearance of pseudo-peak in the time series scattering curve, and these non-heading stage features will interfere with the identification of the heading stage of the crops. The three problems seriously restrict the business application of the radar time series observation in the identification of the heading stage of the crops. It can be seen from the embodiments that the heading stage of the crops in Changxing County of Huzhou City in 2021 is identified, and the precision test result reaches 76.19%. The method provided by the present application effectively tracks the remote sensing time series features of the growth of the crops without relying on prior phenology information, evaluates the periodicity of the multi-round crop planting time series data, determines the dominant periodicity of the annual planting mode, identifies the peak in combination with the growth law of the crops and the cosine component phase information, realizes the identification of the heading stage of the crops in the background of complex and diverse farming practices and frequent cloud and fog weather interference, and can promote the business application of the radar data and provide information support for stable and increased grain production.

[0155] The application embodiment of the present application provides a computer device, which comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the crop heading stage identification method based on radar time series observation.

[0156] The application embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the crop heading stage identification method based on radar time series observation.

[0157] The application embodiment of the present application provides an information data processing terminal, which is used to realize the crop heading stage identification system based on radar time series observation.

[0158] It should be noted that embodiments of the present application can be realized by hardware, software, or a combination of software and hardware. The hardware portion can be realized by a special logic; the software portion can be stored in a memory and executed by a proper instruction execution system, such as a microprocessor or a specially designed hardware. A person of ordinary skill in the art can understand that the above-mentioned apparatus and method can be realized by computer executable instructions and / or included in processor control codes, such as a carrier medium, such as a magnetic disk, CD or DVD-ROM, a programmable memory, such as a read-only memory (firmware), or a data carrier, such as an optical or electronic signal carrier. The apparatus of the present application and its modules can be realized by a hardware circuit, such as a very large scale integrated circuit or a gate array, a semiconductor, such as a logic chip, a transistor, or a programmable hardware device, such as a field programmable gate array, a programmable logic device, or the like, by software executed by various types of processors, or by a combination of the above-mentioned hardware circuit and software, such as firmware.

[0159] The above description is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any modification, equivalent replacement, and improvement within the technical range disclosed by the present application, and within the spirit and principle of the present application, should be covered within the protection scope of the present application.

Claims

1. A method for identifying the heading stage of crops based on radar time-series observation, characterized in that, The depolarization index is obtained through index calculation, the periodic parameters of the time-series depolarization index are extracted through harmonic decomposition, the dominant periodic components of the time-series depolarization index are determined through weight analysis, the heading period of crops is identified through peak detection, and crop heading period is mapped by combining land use products. The specific steps of the method for identifying the heading stage of crops based on radar time-series observations include: S1, Indicator Calculation: Obtain the VH and VV polarization data from annual radar time-series observations, and calculate the depolarization index using the following formula: OF=10 VH / 10 / (10 VH / 10 +10 VV / 10 ) S2, Harmonic decomposition: For the time-series depolarization index D[t], where t is the normalized annual day, harmonic decomposition is performed using the following formula: 'a' represents the time series mean, A i Represents the magnitude of the i-th cosine term. Represents the phase of the i-th cosine term; a and A are obtained using least-squares fitting. i and The value; S3, Weight Analysis: The weight W of the i-th cosine term i Defined as follows: The maximum weight W of the three cosine terms Max for: When the weight W of the i-th cosine term i Equal to the maximum weight W of the three cosine terms Max At that time, that is: IN i =In Max The dominant periodic component is the i-th cosine term; S4, Peak Detection: The first-order cosine term has one peak, and the peak date C1 is calculated by the following formula: The second-order cosine term has two peaks, with peak date C. 2a and C 2b Calculated by the following formula: The third-order cosine term has three peaks, with peak date C. 3a C 3b and C 3c Calculated by the following formula: S5, Crop heading stage mapping: The system acquires land use products and extracts cultivated land areas. It then determines the dominant periodic components pixel by pixel within the cultivated land areas, thereby determining the number and date of peaks. Finally, it determines whether the cultivated land is in the heading stage of crops on a monthly basis, thus enabling crop heading stage mapping.

2. The method for identifying the heading stage of crops based on radar time-series observation as described in claim 1, characterized in that, In S3, harmonic decomposition treats the time series DI as a synthesis of constant components and three-order cosine components. Therefore, each peak in the time series DI is a synthesis of the peaks of the three-order cosine components. The local trend of a peak is approximated by the slope from the trough to the peak. This slope is represented in the cosine component as the ratio of amplitude to order i, which represents the weight of the cosine component. The cosine component with the largest weight describes the dominant periodicity of the time series DI.

3. The method for identifying the heading stage of crops based on radar time-series observation as described in claim 1, characterized in that, In S4, the peak position of the cosine component depends on the phase and order i, and the center point of the cosine component is... The date corresponds to the peak of the second-order cosine component and the trough of the first and third-order cosine components. The dates of the remaining peaks are obtained by stepping through a process with the center date as the initial point and the intervals being integer multiples of half the period of the cosine component.

4. The method for identifying the heading stage of crops based on radar time-series observation as described in claim 1, characterized in that, In S5, the premise of identifying the crop heading period is to know the specific cultivated land area. By performing index calculation, harmonic decomposition, weight analysis and peak detection on a pixel-by-pixel basis, the crop heading period of the cultivated land area is obtained. In order to better show the temporal distribution characteristics of the crop heading period, a monthly mapping strategy is adopted to identify whether the cultivated land pixel is in the crop heading period.

5. A crop heading date identification system based on radar time-series observation, applying the crop heading date identification method based on radar time-series observation as described in any one of claims 1 to 4, characterized in that, include: Index calculation module: used to acquire VH and VV polarization data from annual radar time-series observations and calculate the depolarization index; Harmonic decomposition module: used to perform harmonic decomposition on the timing depolarization index; Weight Analysis Module: Used for performing weight analysis; Peak detection module: used for peak detection of the third-order cosine term; Crop heading stage mapping module: used to obtain land use products and extract cultivated land areas; The dominant periodic components are determined pixel by pixel in the cultivated land area, thereby determining the number and date of peaks; the cultivated land is judged month by month to determine whether the cultivated land is in the heading stage of crops, thus realizing the mapping of crop heading stage.

6. A computer device comprising a memory and a processor, the memory storing a computer program, wherein when the computer program is executed by the processor, the processor performs the steps of the crop heading period identification method based on radar time-series observation as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor performs the steps of the crop heading stage identification method based on radar time-series observation as described in any one of claims 1 to 4.

Images