Method for monitoring quantity of rice planthoppers by using rice canopy spectrum ratio index

Abstract

The invention provides a rice planthopper quantity monitoring method for the field of agricultural pests. Aiming at the problems that rice planthopper harm is hidden, the monitoring difficulty is high and the intelligent level is low, the invention provides a method for monitoring the quantity of rice planthoppers by utilizing a rice canopy spectral reflectivity ratio index. The method is characterized by comprising the following steps: measuring the spectral reflectance of rice after being damaged by rice planthoppers with different insect quantities and rice canopies in a control area which are not damaged by rice planthoppers in three wave band regions, namely 420-509 nm, 600-698 nm and 728-986 nm, calculating the ratio of the spectral reflectance of the rice canopies in the rice planthopper damaged area to the spectral reflectance of the rice canopies in the control area, respectively extracting the average value of the ratio of the reflectance in each wave band region, and calculating the average value of the ratio of the reflectance in each wave band region. Constructing three reflectivity ratio indexes, and establishing a plant hopper quantity monitoring model by taking the three ratio indexes as factors; measuring the canopy spectral reflectance of the plot not damaged by the rice planthoppers and the rice field to be measured, extracting three ratio index values and substituting the three ratio index values into the model so as to estimate the quantity of the rice planthoppers in the rice field to be measured. The method can be used for intelligently monitoring the population size of the rice planthopper.

Description

Technical Field

[0001] The invention belongs to the field of agricultural pests, and in particular relates to a method for monitoring the amount of important agricultural pests, rice planthoppers, by utilizing a rice canopy spectral reflectance ratio index. Background Art

[0002] Rice planthoppers are important pests on rice, which have a great impact on rice yield and quality. When the damage is serious, it will cause rice "louse burn" and the harvest will be lost. The prevention and control of rice planthoppers is indispensable in the rice production process. The determination of the prevention and control time and the target field is the key to effectively control rice planthoppers, but the determination of these two factors requires the support of the monitoring results of rice planthoppers in the field. Therefore, accurate and timely monitoring of rice field planthoppers is particularly important in the prevention and control of rice planthoppers.

[0003] At present, the monitoring of rice planthoppers in the field generally adopts the manual field survey method. This method requires one person to go to the field, bend down and pat the base of the rice stem with his hands to shake the planthoppers on the rice into a white porcelain plate below, and then quickly visually count the number of planthoppers in the plate and report it to another person for record, or take a photo of the insects in the porcelain plate and upload it to the automatic identification and counting system to automatically obtain the insect quantity. This method of patting is labor-intensive and requires certain professional knowledge. The scope of each survey is limited, which can no longer meet the needs of modern agriculture to develop in the direction of automation and intelligence. Therefore, it is particularly important to create a rice planthopper monitoring method with intelligent prospects. Although there are research reports on the correlation between the spectrum of rice leaves and canopies and the degree of damage of brown planthoppers or white-backed planthoppers, there is still a lack of spectral indices and methods that can be used for monitoring the number of various rice planthoppers (brown planthoppers, white-backed planthoppers, etc.), so spectral technology has not yet been used for monitoring rice planthoppers in the field. The spectral measurement of the rice canopy can be completed by using a spectrometer mounted on a drone, and the data collected by the spectrometer can be transmitted remotely and in real time to the system platform via a wireless network. The system uses the established spectral monitoring method to automatically estimate the number of planthoppers in the rice field being monitored in real time. Therefore, by creating a practical spectral index and establishing a spectral monitoring method for rice planthoppers, remote, real-time, and automatic monitoring of rice planthoppers can be achieved. Summary of the invention

[0004] The purpose of the present invention is to provide a method suitable for monitoring the occurrence of rice planthoppers in paddy fields. The method uses the spectral reflectance ratio index of the rice canopy to indirectly characterize the number of rice planthoppers, thereby realizing the monitoring of rice planthoppers in paddy fields.

[0005] The objective of the present invention is achieved by the following method: measuring the spectral reflectance of rice canopies damaged by different numbers of rice planthoppers for a certain period of time and healthy rice canopies not damaged by rice planthoppers in three bands of 420-509 nm, 600-698 nm and 728-986 nm, creating a spectral reflectance ratio index, establishing a rice planthopper quantity monitoring model combined with a multi-spectral reflectance ratio index, then collecting the spectral reflectance of each rice field canopy to be tested, extracting each ratio index value, substituting it into the monitoring model, thereby estimating the insect quantity of planthoppers, and realizing automatic monitoring of the insect quantity of rice field planthoppers.

[0006] Effects of the invention: Compared with the commonly used disc-patting method for visually investigating the number of rice planthoppers, the present invention produces the following beneficial effects: (1) The method is easy to operate, fast and accurate, and can complete the measurement of a large area of ​​rice fields in a short time; (2) The monitoring process is fully digitalized, no manual recording is required, and the traceability of the monitoring results is strong; (3) It is applicable to a wide range of people and can be operated without professional knowledge; (4) The spectral reflectance of the rice canopy can be automatically collected using equipment carried by an unmanned aerial vehicle, and the collected data can be wirelessly transmitted to the system platform and the monitoring results can be automatically estimated, thereby realizing the automation and intelligence of rice field planthopper monitoring. DETAILED DESCRIPTION

[0007] The method of the present invention can be implemented in the following manner, but is not limited to this manner.

[0008] The method for setting up the control plot of rice that is not harmed by rice planthoppers is as follows: Rice that grows in the same way as in the field and is not harmed by rice planthoppers is selected as the control plot of rice.

[0009] Method for measuring the hyperspectral reflectance of rice canopy: The hyperspectral reflectance of rice canopy in each plot was measured 7 days after rice planthoppers migrated into rice. The spectral reflectance of rice canopy in three bands of 420-509 nm, 600-698 nm and 728-986 nm was measured by a spectrometer carried by humans or drones. The rice in the control plot was measured in the same way.

[0010] Method for constructing the spectral reflectance ratio index: divide the spectral reflectance of the rice canopy in the test plot at each band of 420-509, 600-698 and 728-986nm by the spectral reflectance of the rice canopy in the control plot at the corresponding band to obtain the spectral reflectance ratio of the rice canopy in the test plot at each band; divide the area into three bands of 420-509, 600-698 and 728-986, and calculate the average value of the spectral reflectance ratio of all bands in each area, recorded as R1, R2 and R3, as the spectral reflectance ratio index.

[0011] Method for establishing a monitoring model for rice planthoppers: measure the canopy spectral reflectance of more than 5 rice plots, and manually survey the number of rice planthoppers in each plot, and use linear regression and other modeling methods to construct a relationship model between three spectral reflectance ratio indices (R1, R2 and R3) and the number of rice planthoppers (N), such as N=C0+C1R1+C2R2+C3R3, where C0, C1, C2, and C3 are constants, and the values ​​are determined according to the actual conditions such as the rice growth period and the number of days of rice planthopper damage. The relationship model can be selected according to the rice variety and planthopper species composition.

[0012] Monitoring method for rice planthoppers: Measure the spectral reflectance of the rice canopy in the test area and the control area at 420-509, 600-698 and 728-986 nm, extract three spectral reflectance ratio indices, substitute the index values ​​into the relationship model, and the amount of rice planthoppers can be estimated.

[0013] Implementation effect: The insect quantity monitoring effect of brown planthopper and white-backed planthopper was measured during the tillering stage and heading stage of the rice variety "Wuyunjing No. 7" when the rice was damaged for 1-4 weeks. The results showed that the three established spectral reflectance ratio indices had good effects on monitoring the insect quantity of rice planthoppers, as shown below.

[0014] Implementation effect 1: Different numbers of brown planthoppers or white-backed planthoppers were inoculated on rice during the tillering stage. When the rice was damaged for 1, 2, 3, and 4 weeks, the reflectance of the rice canopy spectrum at 420-509, 600-698, and 728-986 nm was measured, respectively. The rice without insect inoculation was used as the control, and the three spectral reflectance ratio index values ​​of rice in each insect inoculation area were calculated to establish a regression model for monitoring the number of brown planthoppers when the rice was damaged for 1-4 weeks. The determination coefficient of the model was between 0.665 and 0.9513, all reaching the extremely significant level (P<0.01), which had a good effect on monitoring the number of brown planthoppers. Although these three ratio indices could not establish a linear model to monitor the number of white-backed planthoppers on rice during the tillering stage for 1 and 2 weeks, they had a good effect on monitoring the number of white-backed planthoppers when the rice was damaged for 3 and 4 weeks. The determination coefficients of the model were 0.486 and 0.706, respectively, reaching the extremely significant level (Table 1).

[0015] Table 1 Monitoring model and effect of spectral reflectance ratio index of insect abundance of brown planthopper and white-backed planthopper on rice during tillering period for 1-4 weeks

[0016] Planthopper species Hazard time (weeks) Monitor Model Model determination coefficient P-value Brown Planthopper 1 <![CDATA[N=57.477-22.988R1+12.540R2-38.101R3]]> 0.9513 <0.0001 2 <![CDATA[N=24.007-7.894R1+9.081R2-19.339R3]]> 0.793 <0.0001 3 <![CDATA[N=15.823+1.039R1+4.076R2-14.923R3]]> 0.854 <0.0001 4 <![CDATA[N=8.418+14.703R1-3.736R2-14.642R3]]> 0.665 <0.0001 White-backed Planthopper 1 - - - 2 - - - 3 <![CDATA[N=21.390+4.614R1-0.041R2-16.903R3]]> 0.486 0.0030 4 <![CDATA[N=15.184+31.319R1-10.864R2-28.592R3]]> 0.706 <0.0001

[0017] Note: - indicates a wireless monitoring model.

[0018] Implementation effect 2: Different numbers of brown planthoppers or white-backed planthoppers were inoculated on rice at the booting stage. When the rice was damaged for 1, 2, 3, and 4 weeks, the reflectance of the rice canopy spectrum at 420-509, 600-698, and 728-986 nm was measured, respectively. The rice without insect inoculation was used as the control, and the three spectral reflectance ratio index values ​​of rice in each insect inoculation area were calculated. Although the linear model of these three ratio indexes could not fit the insect quantity of brown planthoppers when it was damaged for 1 week, it could fit the insect quantity when it was damaged for 2, 3, and 4 weeks well. The determination coefficients of the models were 0.39, 0.34, and 0.466, respectively, all reaching significant levels. These three spectral reflectance ratio indices could linearly fit the insect quantity of white-backed planthoppers on rice at the booting stage for 1, 2, 3, and 4 weeks well. The determination coefficients of the linear model were 0.39, 0.536, 0.571, and 0.532, respectively, all reaching significant levels, and had a good effect in monitoring the insect quantity of white-backed planthoppers (Table 2).

[0019] Table 2 Monitoring model and effect of spectral reflectance ratio index of insect abundance of brown planthopper and white-backed planthopper on rice during booting stage for 1-4 weeks

[0020] Planthopper species Hazard time (weeks) Monitor Model Model determination coefficient P-value Brown Planthopper 1 - - - 2 <![CDATA[N=-113.333+174.508R1-23.953R2-22.568R3]]> 0.390 0.014 3 <![CDATA[N=79.753-25.239R1+4.515R2-47.869R3]]> 0.34 0.031 4 <![CDATA[N=17.954+20.616R1-1.356R2-31.471R3]]> 0.466 0.004 White-backed Planthopper 1 <![CDATA[N=-79.643+114.652R1+3.197R2-25.208R3]]> 0.390 0.014 2 <![CDATA[N=66.795+77.075R1-28.377R2-95.745R3]]> 0.536 0.001 3 <![CDATA[N=35.527+23.073R1+2.254R2-48.506R3]]> 0.571 <0.0001 4 <![CDATA[N=57.061+13.931R1-6.174R2-54.133R3]]> 0.532 0.001

[0021] Note: - indicates a wireless monitoring model.

[0022] The implementation effect of this method on the two types of rice planthoppers, brown planthopper and white-backed planthopper, for more than 1-4 weeks in two growth periods of rice respectively shows that the three spectral reflectance ratio indices R1, R2, and R3 can better characterize the pest of rice planthoppers, and the effect is better when the damage of rice planthoppers is serious, and it can be used for rice planthopper population monitoring. This method only needs to measure the spectral reflectance of the rice canopy. It is simple to operate, has low professional requirements, and low work intensity. It can be completed by one person; at the same time, the spectral reflectance of the rice canopy can be automatically measured by the equipment carried by the drone. Using this method to monitor the size of rice planthoppers has automatic and intelligent prospects.

Claims

1. A method for monitoring the amount of planthoppers in rice fields, characterized in that: The spectral reflectance of the canopy of rice plants damaged by different numbers of rice planthoppers and the control rice plants not damaged by rice planthoppers in three bands of 420-509 nm, 600-698 nm and 728-986 nm was measured, and the ratio of the spectral reflectance of the damaged rice canopy to that of the undamaged rice was calculated. The average values ​​of the reflectance ratios in the three bands were extracted, and three spectral reflectance ratio indices were constructed. The three ratio indices were combined to establish a monitoring model for the number of planthoppers on rice. A control plot not damaged by rice planthoppers was established in the field, and the spectral reflectance of the canopy of the control plot and all the tested rice fields was measured. The three spectral reflectance ratio index values ​​of each rice field were extracted and substituted into the monitoring model, so as to estimate the number of planthoppers in each tested rice field.