A leaf area index fine correction data processing method

Abstract

The application discloses a kind of leaf area index precision correction data processing methods, first, according to plant canopy image calculation LAI data, obtain each node daily high time resolution LAI value, then, determine the accurate time of each day sunrise and sunset, obtain LAI value within t1 minutes before sunrise, within t2 minutes after sunset each day, and respectively calculate early, late representative value, with early morning arithmetic average as the LAI representative system of day, then calculate the confidence of each day LAI value, according to the accuracy requirement of application, select and set the LAI representative value of day above confidence, obtain the LAI time series value with higher confidence.The application eliminates a large number of invalid LAI data, only retains an effective LAI value as the representative value of day, through the confidence test of each day LAI value, obtains the LAI time series value with higher reliability.The application judges sunrise and sunset time, and determines the effective value of day more scientifically and effectively, and is convenient for user to understand and manage LAI data in real time.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of remote sensing information and big data technology, and more specifically relates to a leaf area index data fine correction method, which can be applied to fine correction of leaf area index data indirectly calculated and measured by a wireless sensor network after shooting a canopy image. BACKGROUND

[0002] LAI (Leaf Area Index) is a multiple of the total area of plant leaves per unit of land area, and is an important indicator reflecting the growth of plant populations. The size of LAI is directly related to the final yield, so real-time monitoring and measurement of this parameter have great practical significance in the agricultural industry.

[0003] Studies have shown that the measurement method of plant leaf area index has evolved from the initial destructive direct measurement method to the current image processing method, remote sensing image method, and sensor method. According to actual needs, many advanced devices for measuring leaf area index with high efficiency and high accuracy have been put into use, such as LAI-2000, etc. These devices are measured by manually holding the canopy analysis instrument in the field, which is very inconvenient for long-term observation of the leaf area index of plants in a certain area. Therefore, many scholars have begun to study automatic networking leaf area index measurement methods. Leaf area index data (LAI data) is indirectly calculated and measured by a wireless sensor network after shooting a canopy image or measuring canopy optical information. The automatic measurement system can obtain measurement results every 3-5 minutes at the fastest, and collect a large number of leaf area index measurement values every day. At the same time, due to the influence of natural environment such as light changes and wind disturbances, the measurement values fluctuate greatly. Considering that the daily change of leaf area index is not large, how to extract the representative value of daily leaf area index from these large amounts of leaf area index data has become the focus of work.

[0004] Currently, scholars at home and abroad have made certain achievements in direct and indirect measurement of LAI data, and have also developed many new sensor designs to achieve monitoring of LAI data. However, there is currently no specific data processing method for this large amount of data. Most data processing methods are generally to deal with null values, missing values, regularization, and principal component analysis, etc. For continuously measured LAI data, this large amount of data with high environmental requirements and many invalid data has no specificity.

[0005] The stable window method mentioned in the Chinese invention authorized on April 18, 2023, with announcement number CN 111966952 B and title "A Method for Filtering Leaf Area Index Data," is a relatively effective method for processing this large amount of data. However, the stable window method relies on stable light conditions over a certain period of time, and due to the influence of light transmission and diffraction within the canopy, it is easy to underestimate the measured values. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of current LAI data measurement methods, such as high requirements for the measurement environment, a large amount of invalid data, and the tendency to underestimate the measured values ​​due to the influence of transmission and diffraction within the canopy. This invention proposes a leaf area index (LAI) fine correction data processing method, which more conveniently and scientifically extracts the representative values ​​of the daily LAI values ​​from a large amount of LAI data measured at various nodes, making it easier for users to understand and manage LAI data in real time.

[0007] To achieve the above-mentioned objective, the leaf area index fine-correction data processing method of the present invention is characterized by comprising the following steps:

[0008] (1) Obtain plant canopy images collected at fixed intervals at each node and calculate LAI data to obtain high temporal resolution LAI values ​​for each node;

[0009] (2) Calculate the exact sunrise and sunset times for each day, obtain the LAI value within t1 minutes before sunrise and t2 minutes after sunset, and calculate the representative value of the morning LAI for each day based on the obtained LAI value within t1 minutes before sunrise, and calculate the representative value of the evening LAI for each day based on the obtained LAI value within t2 minutes after sunset.

[0010] (3) Calculate the daily representative value of LAI: Take the average of the daily morning and evening representative values ​​of LAI as the daily representative value of LAI;

[0011] (4) Calculate the confidence level of the daily LAI representative value. Based on the accuracy requirements of the application, select the LAI representative value with a confidence level above the set value to obtain the LAI time series value with a higher confidence level.

[0012] The objective of this invention is achieved as follows:

[0013] This invention presents a leaf area index (LAI) fine-calibration data processing method. First, LAI data is calculated based on plant canopy images to obtain high-temporal-resolution LAI values ​​for each node daily. Then, the precise times of sunrise and sunset are determined daily, yielding LAI values ​​within t1 minutes before sunrise and t2 minutes after sunset each day. Representative values ​​for the morning and evening are calculated separately. Next, the confidence level of each daily LAI value is calculated. Based on the application's accuracy requirements, representative LAI values ​​with a confidence level above a set threshold are selected to obtain a high-confidence LAI time series value. This invention eliminates a large amount of invalid LAI data, retaining only one valid LAI value as the representative value for the day. By verifying the confidence level of each daily LAI value, a high-reliability LAI time series value is obtained. Based on the principle of LAI measurement, it requires diffuse light conditions, such as cloudy days, before sunrise, or after sunset. Therefore, this invention's method of determining sunrise and sunset times and thus identifying the valid value for the day is more scientific and effective, facilitating real-time understanding and management of LAI data for users. Meanwhile, this invention has a superior effect on processing large amounts of leaf area index data collected in the field that vary greatly with the environment, and has great practical significance and wider applicability for the management and monitoring of vegetation growth. Attached Figure Description

[0014] Figure 1 This is a flowchart of a specific embodiment of the leaf area index fine correction data processing method of the present invention;

[0015] Figure 2 yes Figure 1 Flowchart for calculating early and late LAI representative values;

[0016] Figure 3 This refers to the daily representative LAI value data after fine correction in the leaf area index fine correction data processing method of this invention;

[0017] Figure 4 This is a scatter plot of the daily representative LAI values ​​after a confidence test (confidence level is 0.8).

[0018] Figure 5 This is a scatter plot of daily LAI representative values ​​using the stable window method;

[0019] Figure 6 This is a scatter plot of the daily LAI representative value data using the morning and evening method in this invention. Detailed Implementation

[0020] The specific embodiments of the present invention will now be described with reference to the accompanying drawings to enable those skilled in the art to better understand the invention. It should be particularly noted that in the following description, detailed descriptions of known functions and designs that might obscure the main content of the invention will be omitted here.

[0021] Figure 1 This is a flowchart of a specific implementation of the leaf area index fine correction data processing method of the present invention.

[0022] In this embodiment, as Figure 1 As shown, the leaf area index fine-correction data processing method of the present invention includes the following steps:

[0023] Step S1: Obtain the high temporal resolution LAI value for each node.

[0024] Plant canopy images were acquired at fixed intervals at each node, and LAI data were calculated to obtain high temporal resolution LAI values ​​for each node.

[0025] In this embodiment, vegetation canopy images of the monitored area are acquired through a wireless sensor network. Image analysis and environmental variable data processing are used to obtain parameters such as canopy porosity, which are then used to calculate the leaf area index (LAI). In this embodiment, all data collected by node 0801 at Yucheng Station from March 1st to December 31st, 2020, are used. During these 10 months, the LAI sensor automatically analyzes the LAI value every 5 minutes and automatically transmits it back to the server via the wireless sensor network.

[0026] Step S2: Calculate the representative values ​​of morning and evening LAI.

[0027] Step S201: Calculate the exact sunrise and sunset times for each day.

[0028] Calculate the local daily sunrise and sunset azimuth angle θ:

[0029]

[0030] in, δ represents the local latitude, and δ represents the latitude of the subsolar point each day;

[0031] Convert the local sunrise and sunset azimuth angle θ into time t to obtain the accurate sunrise and sunset times (Northern Hemisphere) for each day:

[0032] Precise sunrise time: 6:00-t during the summer half-year and 6:00+t during the winter half-year;

[0033] The exact time of sunset is 18:00+t during the summer half-year and 18:00–t during the winter half-year.

[0034] The exact times of sunrise and sunset each day are known only to us and will not be elaborated upon here.

[0035] Step S202: Calculate the representative LAI values ​​for each morning and evening.

[0036] Obtain the LAI value within t1 minutes before sunrise and t2 minutes after sunset each day. Calculate the representative LAI value for the morning based on the obtained LAI value within t1 minutes before sunrise each day, and calculate the representative LAI value for the evening based on the obtained LAI value within t2 minutes after sunset each day.

[0037] t1 and t2 range from 0 to 30. In this embodiment, both t1 and t2 are set to 10, meaning that the LAI values ​​are obtained within 10 minutes before sunrise and 10 minutes after sunset each day. Then, representative morning and evening LAI values ​​are calculated separately: the average or maximum value within 10 minutes before sunrise and 10 minutes after sunset is calculated, or the first value in the morning and the last value in the evening are used as representative morning and evening LAI values, respectively. In this embodiment, the arithmetic mean of the representative morning and evening LAI values ​​within 10 minutes before sunrise and 10 minutes after sunset is taken as the representative morning and evening LAI values.

[0038] Step S3: Calculate the daily LAI representative value

[0039] The average of the daily morning and evening LAI representative values ​​is used as the daily representative LAI value. In this embodiment, the average value is the arithmetic mean.

[0040] Step S4: Confidence Test

[0041] Calculate the confidence level of the daily representative LAI value. Based on the accuracy requirements of the application, select the LAI representative value with a confidence level above the set value to obtain the LAI time series value with a higher confidence level.

[0042] In this embodiment, the confidence level calculation formula is:

[0043] Confidence level = 1 - |LAI morning -LAI night | / (LAI morning +LAI night )

[0044] Among them, LAI morning This is the daily LAI representative value. night This is the LAI value for each evening.

[0045] Based on the accuracy requirements of the application, a value with a certain confidence level is selected. In this embodiment, the confidence level is set to 0.8.

[0046] Specifically, in this embodiment, the high temporal resolution LAI value of node 0801 at Yucheng Station is first obtained. This involves collecting plant canopy images of node 0801 at fixed time intervals each day (5 minutes) and calculating LAI data, resulting in multiple fixed-interval LAI values ​​for node 0801 each day. Then, a representative value for each day's LAI is calculated. Based on the latitude and longitude coordinates of node 0801, the sunrise and sunset times are determined, and LAI values ​​are selected within 10 minutes before sunrise and 10 minutes after sunset each day. Representative values ​​for the morning and evening are calculated separately for each day. In this invention, the representative value for the day's LAI is the average of the morning and evening representative values. Finally, a confidence test is performed, and values ​​with a confidence level of 0.8 or higher are selected as standard values ​​to obtain the LAI time series value of the vegetation at node 0801.

[0047] In this embodiment, Figure 3 The image shows a scatter plot of the daily representative LAI values ​​of all LAI data collected from March 1 to December 31, 2020 at node 0801 of Yucheng Station, after processing by this invention. Only one valid LAI value is retained in the figure as the representative value of the LAI value for that day. Due to the specificity of this LAI data source, a large number of invalid LAI data have been removed. Figure 4 The scatter plot of the daily representative LAI values ​​after confidence level testing (confidence level is 0.8) makes the data results more reliable and facilitates further analysis and utilization of the site LAI data by researchers. Figure 5 This is a scatter plot of daily LAI representative values ​​using the stationary window method. Figure 6 This is a scatter plot of the daily LAI representative value data of the morning and evening method in this invention. It can be clearly seen from the data in the figure that the daily LAI representative value data of the stationary window method fluctuates greatly, while the daily LAI representative value data of the morning and evening method used in this invention varies less, indicating that the data results of the morning and evening method used in this invention are more real and reliable.

[0048] Although the illustrative specific embodiments of the present invention have been described above to enable those skilled in the art to understand the invention, it should be understood that the invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the invention as defined and determined by the appended claims, and all inventions utilizing the concept of the present invention are protected.

Claims

1. A method for processing leaf area index (LAI) fine-calibration data, characterized in that, Includes the following steps: (1) Obtain plant canopy images collected at fixed intervals at each node and calculate LAI data to obtain high temporal resolution LAI values ​​for each node; (2) Calculate the exact sunrise and sunset times for each day, obtain the LAI value within t1 minutes before sunrise and t2 minutes after sunset, and calculate the representative value of the morning LAI for each day based on the obtained LAI value within t1 minutes before sunrise, and calculate the representative value of the evening LAI for each day based on the obtained LAI value within t2 minutes after sunset. (3) Calculate the daily representative value of LAI: Take the average of the daily morning and evening representative values ​​of LAI as the daily representative value of LAI; (4) Calculate the confidence level of the daily LAI representative value. Based on the accuracy requirements of the application, select the LAI representative value with a confidence level above the set value to obtain the LAI time series value with a higher confidence level.

2. The leaf area index fine-correction data processing method according to claim 1, characterized in that, The values ​​of t1 and t2 range from 0 to 30.

3. The leaf area index fine-correction data processing method according to claim 1, characterized in that, Obtain the LAI values ​​within 10 minutes before sunrise and 10 minutes after sunset each day, and then calculate the representative LAI values ​​for the morning and evening respectively: calculate the average or maximum value within 10 minutes before sunrise and 10 minutes after sunset, or use the first value in the morning and the last value in the evening as the representative LAI values ​​for the morning and evening respectively.

4. The leaf area index fine-correction data processing method according to claim 1, characterized in that, The arithmetic mean of the representative values ​​of morning and evening LAI within 10 minutes before sunrise and after sunset is taken as the average value of morning and evening LAI.

5. The leaf area index fine-correction data processing method according to claim 1, characterized in that, The confidence level calculation formula is as follows: Confidence level = 1 - |LAI morning -LAI night | / (LAI morning +LAI night ) Among them, LAI morning This is the daily LAI representative value. night This is the LAI value for each evening.

6. The leaf area index fine-correction data processing method according to claim 1, characterized in that, The set confidence level is 0.8.

Images