A method and device for monitoring the planting situation of a paddy field
By coupling time-series radar data with multiple features of rice, the problems of cloud and fog interference and low identification accuracy in rice planting monitoring have been solved, achieving efficient and low-cost monitoring of rice planting conditions and ensuring accurate management of paddy fields.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SURVEYING & MAPPING INST LANDS & RESOURCE DEPT OF GUANGDONG PROVINCE
- Filing Date
- 2022-11-24
- Publication Date
- 2026-06-12
Smart Images
Figure CN115902871B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of remote sensing crop identification technology, specifically to a method and device for monitoring paddy field planting conditions based on the coupling of time-series radar data with multiple features of rice. Background Technology
[0002] Arable land is an essential resource for ensuring food security. However, with population growth and improved living standards, the area and quantity of arable land are decreasing year by year. Therefore, it is urgent to implement the strictest arable land protection system and firmly safeguard the 1.8 billion mu (120 million hectares) red line for arable land. Paddy fields account for about a quarter of the country's total arable land area. To curb the "non-agriculturalization" and prevent the "non-grainization" of arable land, various regions have promoted the construction of paddy field reclamation projects, helping to resolve the dilemma of balancing paddy field occupation and compensation. Conducting dynamic monitoring and evaluation of paddy field rice planting to ensure that at least one crop of rice is planted in paddy fields each year is essential. Timely, rapid, and accurate information on paddy field rice planting allows for understanding the rice planting situation within the paddy field area, assessing the intensity of paddy field utilization, and estimating the total rice planting area. This provides a reference for government management and agricultural policy formulation, and is also a requirement for my country's food security strategy.
[0003] The existing commonly used methods for monitoring rice planting conditions are based on medium- and high-resolution optical remote sensing data, mainly including two categories of methods: (1) Rice identification method based on single-phase remote sensing images, which selects single-phase images of key growth periods of rice and uses the spectral information of the images to identify rice; (2) Rice identification method based on multi-phase remote sensing data, which selects multiple phase images during the growth period, generates time-series curves, and analyzes the spectral characteristics of rice at different periods to identify rice planting conditions.
[0004] The existing methods for monitoring rice cultivation conditions have the following main drawbacks:
[0005] (1) Rice identification based on optical images has rich spectral information, but the acquisition of optical images is easily affected by weather such as clouds and fog, and high spatial resolution and high temporal resolution cannot be obtained at the same time; (2) When rice is identified based on single-temporal remote sensing images, the rice information obtained is insufficient, and there will be situations of "same object, different spectrum" and "same spectrum, different object"; (3) Rice identification based on multi-temporal remote sensing data can be carried out by taking advantage of the significant spectral differences in different growth stages of crops, but it has high requirements for image quality, and cloudy weather and noise will affect its identification accuracy. Summary of the Invention
[0006] To address the shortcomings of existing technologies, the present invention aims to provide a monitoring technology and method for paddy field planting based on the coupling of time-series radar data with multiple features of rice, thereby solving practical problems such as high cost, low efficiency, and long time consumption in paddy field management, as well as monitoring problems such as insufficient rice identification information and low accuracy.
[0007] To achieve the above objectives, the present invention adopts the following technical solution:
[0008] In a first aspect, the present invention provides a method for monitoring paddy field planting conditions, comprising:
[0009] Data preparation steps:
[0010] Acquire temporal radar data and rice samples;
[0011] The acquired time-series radar data is preprocessed to obtain preprocessed time-series radar data.
[0012] The standard time series curve of the rice sample area is extracted from the preprocessed time-series radar data to obtain the time-series curve of the rice sample.
[0013] Steps for extracting key information about rice:
[0014] Based on rice sample data, we determined the characteristics of water and rice transplanting, the VH value range of rice heading signal, and the date difference between transplanting and heading. These three indicators were used as phenological characteristics for rice identification.
[0015] The statistical characteristics of time series data use variance.
[0016] Similarity metric calculation steps:
[0017] Based on the preprocessed temporal radar data, the Euclidean distance measure and spectral similarity measure between the temporal curve of each pixel in the image and the temporal curve of the rice sample are calculated, and then the spectral measurement value of the rice is calculated.
[0018] The steps for coupled multi-feature recognition of rice:
[0019] By utilizing rice sample data, the optimal thresholds for identifying rice phenological characteristics and time-series statistical features were determined. A decision tree method was then employed to couple spectral measurements, phenological characteristics, and time-series statistical features of rice to identify rice planting conditions within paddy fields.
[0020] Furthermore, we use rice transplanting date information to eliminate areas without water; we use heading date information to eliminate water bodies; and we use the date difference between transplanting and heading dates to eliminate grasslands growing in water.
[0021] Furthermore, the time-series radar data is time-series Sentinel-1A image data within the year.
[0022] Furthermore, the rice samples were obtained in the following manner:
[0023] Field surveys were conducted to take photos of the paddy fields, recording the location of the photos and the azimuth angle, and obtaining rice samples.
[0024] Furthermore, the Euclidean distance measurement is calculated as follows:
[0025]
[0026] Normalizing the above formula to 0-1, we get the following formula:
[0027] E d =(Ed orig -m) / (Mm)
[0028] In the formula: m and M are Ed orig The minimum and maximum values are defined as p and t, respectively, which are the sample dataset vector and the monitoring dataset vector of rice. n is the dimension of the sample data, i.e., the number of image periods used.
[0029] Furthermore, the spectral similarity measure is calculated as follows:
[0030]
[0031] In the formula: μ and σ are the mean and standard deviation of two vectors (t and p), respectively, and t, p, and n are the same as above.
[0032] Furthermore, the spectral measurement values are calculated as follows:
[0033] Spectral measurements are a combination of Euclidean distance and spectral correlation measures.
[0034]
[0035] A threshold of 1 is set to identify areas with spectral measurement values less than 1 as rice-grown areas and areas with values greater than 1 as non-rice-grown areas.
[0036] Furthermore, the optimal thresholds for identifying rice phenological characteristics and time-series statistical characteristics are as follows:
[0037] The VH value at the rice transplanting stage is less than or equal to -17; the VH value at the rice heading stage is between -19 and -12; the difference between the transplanting and heading stages is 45 to 90 days; and the variance is greater than 1.
[0038] A rice-growing area is one that meets all of the above threshold conditions.
[0039] In a second aspect, the present invention provides a paddy field planting monitoring device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.
[0040] Thirdly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the methods described above.
[0041] The beneficial effects of this invention are as follows:
[0042] Radar data imaging is unaffected by clouds and fog, enabling rapid acquisition of effective image data covering large areas and providing full coverage. It features high resolution, all-weather operation, and effective identification of camouflage and penetration of concealment. Studying different characteristic information of rice allows for more accurate rice identification. Compared with existing technologies, this invention, based on time-series radar data, is unaffected by weather conditions such as clouds and fog. It leverages the significant differences exhibited by crops at different growth stages to obtain rich information about rice growth, facilitating the extraction of unique key information and reducing instances of "different species with the same characteristics" and "different species with the same characteristics." This invention enriches rice identification information and improves the accuracy of rice planting identification by addressing key phenological and statistical information extraction and the setting of multi-feature coupling thresholds. It also offers advantages such as high efficiency, short processing time, and low cost in later management.
[0043] By coupling time-series radar data with multiple rice characteristics, the planting situation in paddy fields is monitored, ensuring the improvement of the dynamic monitoring system for arable land protection. The application of data from paddy field planting monitoring can curb the "non-agriculturalization" of arable land and prevent its "non-grainization." As investment in paddy fields, as part of agriculture, rural areas, and farmers, gradually increases, and under the guidance of pro-farmer policies such as direct grain subsidies, improved seed subsidies, and comprehensive subsidies, the situation is stabilizing. The government can formulate policies and implement management actions based on monitoring data to ensure that paddy fields are not abandoned or neglected, guaranteeing safe planting of grain, which is of great significance to national food security and social stability. Attached Figure Description
[0044] Figure 1 This is a detailed implementation roadmap of the paddy field planting monitoring method based on the coupling of time-series radar data and multiple features of rice provided in Embodiment 1 of the present invention.
[0045] Figure 2 This is a schematic diagram of the composition of the paddy field planting monitoring device based on the coupling of time-series radar data and multiple features of rice provided in Embodiment 2 of the present invention. Detailed Implementation
[0046] The present invention will now be further described with reference to the accompanying drawings and specific embodiments:
[0047] See Figure 1 As shown in the figure, the paddy field planting monitoring method based on the coupling of time-series radar data and multiple features of rice provided in this embodiment specifically includes the following steps:
[0048] (1) Data preparation steps:
[0049] 1) Time-series Sentinel-1A images for the year were obtained from the European Space Agency's website. The Sentinel-1A images used for monitoring are L1-level ground distance wide-swath interferometric data with VH polarization, a swath width of 250 km, a distance resolution of 5 m, an azimuth resolution of 20 m, and a spatial resolution of 10 m after preprocessing. Field surveys were conducted to take photos of the paddy fields, record the location of the photo points and the azimuth angle, and obtain rice sample data.
[0050] 2) The radar data was preprocessed using SNAP software developed by the European Space Agency. The data preprocessing mainly included six aspects: thermal noise elimination, orbit correction, radiometric calibration, refined Lee filtering, range-Doppler terrain correction, and conversion of linear scale to logarithmic scale (dB).
[0051] 3) An algorithm written in Python is used to automatically extract the standard time series curve of the rice sample area on the Sentinel-1A time series image. The nodes on the time series curve are the image dates, and the values are the average backscattering coefficients of the sample blocks, thus obtaining the time series curve of the rice samples.
[0052] (2) Steps for extracting key information about rice:
[0053] 1) Understanding the rice growth cycle: Rice differs from other crops in that it requires transplanting in a water-soil mixed environment. Because rice growth requires irrigation, the specular scattering effect from the water surface reduces the backscattering coefficient. After transplanting in paddy fields, due to the interaction between radar signals and water and rice, double scattering (surface scattering and volume scattering) becomes dominant. As rice grows, this increases the surface roughness of the paddy field, leading to an increase in the backscattering coefficient, reaching its maximum at the heading stage. Therefore, water and rice transplanting characteristics, and rice heading signals are key phenological features of rice. Based on these key phenological features, the VH value range of water and rice transplanting characteristics, rice heading signals, and the date difference between transplanting and heading stages are determined using rice sample data. These three indicators are used as phenological information for rice identification. The rice transplanting period corresponds to the minimum value of the rice growth curve, with a VH value generally less than or equal to -17; the rice heading period corresponds to the maximum value of the rice growth curve, with a VH value between -19 and -12; the date difference between the transplanting and heading periods ranges from 45 to 90 days. The rice transplanting period information is used to eliminate areas without water; the heading period information is used to eliminate areas with water bodies; and the date difference between the transplanting and heading periods is used to eliminate grasslands growing in water.
[0054] 2) The time series statistical feature uses variance. Variance can measure how much a set of numbers differs from its mean. It can be used as an indicator for outlier removal. The VH value of rice changes rapidly during the growing season. It is expected that the variance value of rice's VH will be higher than that of other crop types. Areas with a variance less than 1 are removed to distinguish rice from land cover with smaller changes.
[0055] (3) Similarity metric calculation steps
[0056] Based on the preprocessed intra-year time-series radar satellite imagery, the Euclidean distance measure and spectral similarity measure between the time-series curve of each pixel in the image and the time-series curve of the rice sample are calculated. Then, the spectral measurement values are calculated, and finally, the spectral measurement result map is obtained. The spectral measurement value is greater than zero, and the smaller the value, the greater the similarity.
[0057] Euclidean distance is a metric used to measure the separation or proximity of data samples. The Euclidean distance is defined as:
[0058]
[0059] Normalizing the above formula to 0-1, we get the following formula:
[0060] E d =(Ed orig -m) / (Mm)
[0061] In the formula: m and M are Ed orig The minimum and maximum values are defined as p and t, respectively, which are the sample dataset vector and the monitoring dataset vector of rice. n is the dimension of the sample data, i.e., the number of image periods used.
[0062] Spectral correlation measures, such as the correlation between time series curves of rice and other land cover types, can be used as a similarity measure.
[0063]
[0064] In the formula: μ and σ are the mean and standard deviation of two vectors (t and p), respectively, and t, p, and n are the same as above.
[0065] Spectral measurements are a combination of Euclidean distance and spectral correlation measures.
[0066]
[0067] According to the design scheme verification, a threshold of 1 was set, and areas with spectral measurement values less than 1 were identified as rice-grown areas, while areas with values greater than 1 were identified as non-rice-grown areas.
[0068] (4) Steps for coupling multiple features in rice identification
[0069] 1) After determining the optimal thresholds for each feature using rice sample data, a decision tree method is used to couple multiple rice features. The union of each feature is then determined, satisfying the following conditions: ① spectral measurement value less than 1; ② VH value at the rice transplanting stage less than or equal to -17; ③ VH value at the rice heading stage between -19 and -12; ④ the date difference between the transplanting and heading stages is between 45 and 90 days; ⑤ variance greater than 1. These conditions are used to identify rice-grown areas and non-rice-grown areas.
[0070] 2) The accuracy of the monitoring results was evaluated using field survey data, and the accuracy rate of rice planting identification at the provincial level was 86.16%.
[0071] In summary, the paddy field planting monitoring method based on the coupling of time-series radar data and multiple features of rice provided in this embodiment has the following technical advantages compared with the prior art:
[0072] 1. By using time-series radar remote sensing images coupled with rice multi-feature coupling method to monitor paddy field planting conditions, we can achieve efficient paddy field management, reduce traditional field inspections, interpretation, and photography, reduce labor costs, time and money expenditures, and streamline government funding for paddy field monitoring.
[0073] 2. The multi-temporal Sentinel-1A radar data used has a short satellite revisit cycle and the data acquisition is not affected by weather such as clouds and fog. It can acquire large-area, full-coverage effective image data for free and quickly, saving a lot of image purchase costs.
[0074] 3. Strengthen the protection and management of natural resources and arable land. By monitoring the planting situation of paddy fields, we can extend the monitoring to rice planting. Based on the ecological environment and geographical conditions of various regions, we can formulate scientific and reasonable planting plans, reduce operating costs, and create greater economic benefits.
[0075] Example 2:
[0076] See Figure 2 As shown, the paddy field planting monitoring device provided in this embodiment includes a processor 21, a memory 22, and a computer program 23 stored in the memory 22 and executable on the processor 21, such as a paddy field planting monitoring program. When the processor 21 executes the computer program 23, it implements the steps of Embodiment 1 described above, for example... Figure 1 The steps are shown.
[0077] For example, the computer program 23 can be divided into one or more modules / units, which are stored in the memory 22 and executed by the processor 21 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program 23 in the paddy field planting monitoring device.
[0078] The paddy field planting monitoring device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The paddy field planting monitoring device may include, but is not limited to, a processor 21 and a memory 22. Those skilled in the art will understand that... Figure 2 This is merely an example of a paddy field planting monitoring device and does not constitute a limitation on the paddy field planting monitoring device. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the paddy field planting monitoring device may also include input / output devices, network access devices, buses, etc.
[0079] The processor 21 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.
[0080] The memory 22 can be an internal storage unit of the paddy field planting monitoring device, such as the hard drive or memory of the device. The memory 22 can also be an external storage device, such as a plug-in hard drive, SmartMediaCard (SMC), SecureDigital (SD) card, or FlashCard. Furthermore, the memory 22 can include both internal and external storage units. The memory 22 is used to store the computer program and other programs and data required by the paddy field planting monitoring device. The memory 22 can also be used to temporarily store data that has been output or will be output.
[0081] Example 3:
[0082] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in Embodiment 1.
[0083] The computer-readable medium shown can be any means that can contain, store, communicate, propagate, or transmit a program for use in or in conjunction with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, for example, by optically scanning the paper or other medium, then editing, interpreting, or otherwise processing it as necessary to obtain the program electronically, and then storing it in computer memory.
[0084] The above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made based on the essence of the content of the present invention should be covered within the scope of protection of the present invention.
[0085] For those skilled in the art, various other corresponding changes and modifications can be made based on the technical solutions and concepts described above, and all such changes and modifications should fall within the protection scope of the claims of this invention.
Claims
1. A method for monitoring paddy field planting conditions, characterized in that, include: Data preparation steps: Acquire temporal radar data and rice samples; The acquired time-series radar data is preprocessed to obtain preprocessed time-series radar data. The standard time series curve of the rice sample area is extracted from the preprocessed time-series radar data to obtain the time-series curve of the rice sample. Steps for extracting key information about rice: Based on rice sample data, we determined the characteristics of water and rice transplanting, the VH value range of rice heading signal, and the date difference between transplanting and heading. These three indicators were used as phenological characteristics for rice identification. The statistical characteristics of time series data use variance. Similarity metric calculation steps: Based on the preprocessed temporal radar data, the Euclidean distance measure and spectral similarity measure between the temporal curve of each pixel in the image and the temporal curve of the rice sample are calculated, and then the spectral measurement value of the rice is calculated. The steps for coupled multi-feature recognition of rice: By using rice sample data, the optimal thresholds for identifying rice phenological characteristics and time series statistical characteristics were determined. The decision tree method was used to couple the spectral measurements, phenological characteristics, and time series statistical characteristics of rice to identify the rice planting situation in paddy fields.
2. The method for monitoring paddy field planting conditions as described in claim 1, characterized in that, Use rice transplanting date information to eliminate areas without water; use heading date information to eliminate water bodies; use the date difference between transplanting and heading dates to eliminate grasslands growing in water.
3. The method for monitoring paddy field planting conditions as described in claim 1, characterized in that, The time-series radar data refers to the time-series Sentinel-1A imagery data within the year.
4. The method for monitoring paddy field planting conditions as described in claim 1, characterized in that, The rice samples were obtained in the following manner: Field surveys were conducted to take photos of the paddy fields, recording the location of the photos and the azimuth angle, and obtaining rice samples.
5. The method for monitoring paddy field planting conditions as described in claim 1, characterized in that, The Euclidean distance measurement is calculated as follows: Normalizing the above formula to 0-1, we get the following formula: In the formula: m and M are Ed orig The minimum and maximum values are defined as p and t, respectively, which are the sample dataset vector and the monitoring dataset vector of rice. n is the dimension of the sample data, i.e., the number of image periods used.
6. The method for monitoring paddy field planting conditions as described in claim 5, characterized in that, The spectral similarity measure is calculated as follows: In the formula: μ and σ are the mean and standard deviation of two vectors t and p, respectively.
7. The method for monitoring paddy field planting conditions as described in claim 6, characterized in that, The method for calculating the spectral measurement values is as follows: Spectral measurements are a combination of Euclidean distance and spectral correlation measures. ； A threshold of 1 is set to identify areas with spectral measurement values less than 1 as rice-grown areas and areas with values greater than 1 as non-rice-grown areas.
8. The method for monitoring paddy field planting conditions as described in claim 7, characterized in that, The optimal thresholds for identifying rice phenological characteristics and time-series statistical characteristics are as follows: The VH value at the rice transplanting stage is less than or equal to -17; the VH value at the rice heading stage is between -19 and -12; the difference between the transplanting and heading stages is 45 to 90 days; and the variance is greater than 1. A rice-growing area is one that meets all of the above threshold conditions.
9. A device for monitoring paddy field planting conditions, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 8.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 8.