Water-saving agricultural irrigation system and method

Abstract

The application discloses a water-saving agricultural irrigation system and method, and relates to the technical field of agricultural irrigation. The system comprises an irrigation device, a support frame, a pest situation monitoring device, a control center and a server. The irrigation device is arranged at the bottom of the support frame, and the pest situation monitoring device is fixed to the support frame away from the irrigation device. The server is used for running the water-saving agricultural irrigation method according to the received crop image, and generating irrigation control instructions and pest extermination instructions corresponding to the target field area. The control center is also used for receiving the irrigation control instructions of the server, controlling the opening and closing of the irrigation valve, and controlling the running state of the irrigation device, so that the target field area is irrigated by the irrigation device. The control center is also used for receiving the pest extermination instructions of the server, controlling the running of the ultraviolet light, controlling the running state of the pest situation monitoring device, trapping pests by the pest situation monitoring device, and exterminating the pests. The application reduces the cost and equipment management difficulty of the irrigation device and the pest situation monitoring device.

Description

Technical Field

[0001] This invention relates to the technical field of agricultural irrigation, and in particular to a water-saving agricultural irrigation system and method. Background Technology

[0002] In agricultural production, pests and diseases are a significant factor affecting crop yield, quality, and overall economic benefits, posing a considerable threat and thus requiring control. Simultaneously, to protect the agricultural production environment and improve agricultural product quality, the application of green pest and disease control technologies should be strengthened. Green control refers to the use of green plant protection methods. In related research, green control is also known as green pest control methods. According to my country's definition of plant protection, the concept of green control means that plant protection work should ensure harmonious coexistence between humans and nature, emphasizing the guarantee and support role of high yield, high quality, high efficiency, ecological sustainability, and safety. In the process of agricultural pest and disease control, the traditional method is chemical pesticide control, which is quite effective, but its application can lead to pesticide resistance in pests and diseases.

[0003] Existing agricultural irrigation devices and pest monitoring devices are often set up separately and operate independently, resulting in high equipment costs and difficulties in equipment management.

[0004] Therefore, how to reduce the cost and management difficulty of irrigation devices and pest monitoring devices has become an urgent technical problem to be solved. Summary of the Invention

[0005] The technical problem solved by this invention is that existing irrigation devices and insect monitoring devices are costly and difficult to manage.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a water-saving agricultural irrigation system, the system comprising: an irrigation device, a support frame, an insect monitoring device, a control center, and a server; the irrigation device is located at the bottom of the support frame, the insect monitoring device is fixed to the support frame away from the irrigation device, and the control center is fixed to the support frame; the irrigation device is equipped with an ultraviolet lamp and a camera device, and the irrigation device is equipped with an irrigation valve; the control center is used to send crop images captured by the camera device to the server; the server is used to run a water-saving agricultural irrigation method based on the received crop images, and generate irrigation control instructions and insect control instructions corresponding to the target field area; the control center is also used to receive irrigation control instructions from the server, and control the operation of the irrigation device by controlling the opening and closing of the irrigation valve, so as to irrigate the target field area through the irrigation device; the control center is also used to receive insect control instructions from the server, and control the operation of the insect monitoring device by controlling the operation of the ultraviolet lamp, so as to trap and kill pests through the insect monitoring device.

[0007] The present invention also provides the following technical solution: a water-saving agricultural irrigation method, the method comprising: acquiring multiple crop images of multiple target fields; determining the crop growth stage in the corresponding target field based on the crop images; identifying pests based on the crop images to determine the current pest situation data for each target field; determining the social structure of each type of pest based on the current pest situation data; determining the pesticide resistance corresponding to each pest from a preset table of multiple pesticide resistance structures based on the social structure of each type of pest; and determining the pesticide ratio and pesticide-water ratio for irrigation of each target field based on the pesticide resistance of each type of pest and the crop growth stage.

[0008] In a preferred embodiment of the method described in this invention, before determining the pesticide ratio and pesticide-water ratio for irrigation of each target field based on the pesticide resistance of each type of pest and the crop growth stage, the method further includes: predicting future pest data for each target field based on the current pest data and historical pest data of the multiple target fields; the determination of the pesticide ratio and pesticide-water ratio for irrigation of each target field based on the pesticide resistance of each type of pest and the crop growth stage includes: determining the pesticide ratio and pesticide-water ratio for irrigation of each target field based on the pesticide resistance of each type of pest, the crop growth stage, and the future pest data of each target field.

[0009] In a preferred embodiment of the method described in this invention, the step of predicting future insect pest data for each target field based on the current insect pest data and the historical insect pest data of the multiple target fields includes: constructing a random variable X = (X1, X2, ..., Xn) based on the current insect pest data and the historical insect pest data of the multiple target fields, where Xi is a three-dimensional matrix, data is at time i, i and n are both positive integers, and i ≤ n; extracting a feature vector Zi for each Xi in the random variable X to obtain a feature variable Z = (Z1, Z2, ..., Zn); performing vector autoregression analysis on the feature variable Z to obtain the feature vector Zn+1 corresponding to the predicted time n+1; mapping the feature vector Zn+1 back to a three-dimensional matrix to obtain a three-dimensional matrix Xn+1 corresponding to time n+1, where Xn+1 is used to characterize the insect pest data corresponding to time n+1.

[0010] In a preferred embodiment of the method described in this invention, the step of extracting a feature vector Zi for each Xi in the random variable X to obtain the feature variable Z = (Z1, Z2, ..., Zn) includes: Based on the principal component analysis method, the feature vector Zi is extracted from each Xi in the random variable X to obtain the feature variable Z = (Z1, Z2, ..., Zn).

[0011] In a preferred embodiment of the method described in this invention, two dimensions of the three-dimensional matrix Xi are used to characterize the positional coordinates between each target field area, one dimension of the three-dimensional matrix Xi is used to characterize the pest category, and the element xpjk in the three-dimensional matrix Xi is used to characterize the quantity data of the k-th type of pest at coordinates (p, j).

[0012] In a preferred embodiment of the method described in this invention, the pest data includes pest type, quantity, and developmental stage; the quantity data of the k-th type of pest in the element xpjk is determined based on the quantity and developmental stage of the k-th type of pest at the (p, j) coordinate.

[0013] In a preferred embodiment of the method described in this invention, the method further includes: determining the quantity coefficient of the k-th type of pest in the (p, j) coordinate based on the proportion of each developmental stage in the social structure of the k-th type of pest in the (p, j) coordinate; and determining the quantity data of the k-th type of pest in the (p, j) coordinate based on the quantity and data coefficient of the k-th type of pest in the (p, j) coordinate.

[0014] In a preferred embodiment of the method described in this invention, the drug resistance structure comparison table includes a comparison relationship between drug resistance multiples and the social structure of pests, and as the drug resistance multiples in the drug resistance structure comparison table increase from low to high, the difference between adjacent drug resistance multiples increases.

[0015] In a preferred embodiment of the method described in this invention, the step of determining the pesticide ratio and pesticide-water ratio for irrigation of each target field based on the pesticide resistance of each type of pest and the crop growth stage includes: determining the effectiveness of the active ingredient in the pesticide based on the pesticide resistance of each type of pest; and determining the pesticide ratio and pesticide-water ratio for irrigation of each target field based on the effectiveness of each active ingredient.

[0016] The beneficial effects of this invention are as follows: by placing the irrigation device at the bottom of the support frame, fixing the insect monitoring device away from the irrigation device to the support frame, and fixing the control center to the support frame, the equipment cost and management difficulty of the irrigation device and the irrigation device are reduced. Attached Figure Description

[0017] Figure 1 This is a basic structural block diagram of a water-saving agricultural irrigation system provided in an embodiment of the present invention; Figure 2 A schematic diagram of the basic process of a water-saving agricultural irrigation method provided in this embodiment of the invention.

[0018] The reference numerals in the detailed embodiments are as follows: 1. Agricultural irrigation system; 11. Control center; 12. Support frame; 13. Ultraviolet lamp; 14. Camera device. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] Example 1, referring to Figure 1 According to one embodiment of the present invention, a water-saving agricultural irrigation system is provided. The agricultural irrigation system 1 includes: an irrigation device, a support frame 12, an insect monitoring device, a control center 11, and a server. The irrigation device is located at the bottom of the support frame 12, the insect monitoring device is fixed to the support frame 12 away from the irrigation device, and the control center 11 is fixed to the support frame 12. The irrigation device is equipped with an ultraviolet lamp 13 and a camera device 14, and the irrigation device is also equipped with an irrigation valve. The control center 11 is used to send crop images captured by the camera device 14 to the server. The server is used to run a water-saving agricultural irrigation method based on the received crop images and generate irrigation control instructions and insect control instructions corresponding to the target field area. The control center 11 is also used to receive irrigation control instructions from the server and control the operation of the irrigation device by controlling the opening and closing of the irrigation valve, so as to irrigate the target field area through the irrigation device. The control center 11 is also used to receive insect control instructions from the server and control the operation of the insect monitoring device by controlling the operation of the ultraviolet lamp 13, so as to trap and kill pests through the insect monitoring device.

[0021] Figure 1 The illustrated agricultural irrigation system 1 is a physical structure with a separate irrigation device. If an insect monitoring device is to be installed, the top of the folded support frame 12 can be pulled out and the insect monitoring device can be fixed to the support frame 12. Furthermore, the irrigation device and the insect monitoring device can be integrated.

[0022] The server and control center 11 can communicate remotely to send and receive data.

[0023] In studying the resistance of pests to pesticides, existing technologies often ignore the migratory characteristics of pests. For example, Chinese patent with publication number "CN115358883B" discloses a method and system for dynamic monitoring of cotton aphid resistance and recommendation of pesticides for control, which characterizes the resistance information of cotton aphids through changes in the insect infestation index.

[0024] However, the above-mentioned method of characterizing cotton aphid resistance information through changes in insect infestation index is prone to misjudging the resistance of pests in actual use because it ignores the mobility of pests migrating between multiple fields. This can lead to incorrect pesticide application to the fields and fail to effectively defend against pests, especially in drier areas where multiple fields are concentrated near water sources and pests can easily migrate between them.

[0025] Based on this, this application provides a water-saving agricultural irrigation method. After the server acquires the image data collected by the insect monitoring device, it runs the water-saving agricultural irrigation method, which includes steps S110 to S170.

[0026] S110: Acquire multiple crop images of multiple target field areas.

[0027] S120, determine the growth stage of crops in the corresponding target field area based on the crop image.

[0028] Image and spectral information is extracted from hyperspectral images of crops. A continuous projection algorithm is used to process the spectral information and select a set of characteristic wavelengths. Spectral features are generated from this set. The continuous projection method selects variables with the least redundant information and those with the least collinearity from the information variables. Subsets are established for each wavelength combination, and the root mean square error (RMSE) of each subset is calculated using multiple linear regression. The subset with the smallest stable RMS error is selected as the feature wavelength subset. Texture and contour features are obtained from the image information using the gray-level co-occurrence matrix. Generally, features such as energy, entropy, contrast, uniformity, correlation, variance, mean, and variance are used to reflect texture and contour features. Typical features of leaf regions corresponding to each growth stage are obtained using crop plant species information. The deviation between the texture and contour features and the typical features of each growth stage is calculated, and the growth stage with the lowest deviation is selected as the growth stage of the crop plant. The average growth stage of the crop within the target area is determined based on the growth stage of the crop plant. Specifically, refer to Chinese Patent No. CN115358883A for a more detailed implementation of S110.

[0029] S130 identifies pests based on crop images and determines the current pest situation data for each target field area.

[0030] The pest data includes the type, quantity, and developmental stage of the pests.

[0031] S140, determine the social structure of each type of pest based on current pest data.

[0032] Deep learning models (YOLO or UNet++) are used to segment insect-damaged areas and extract morphological features (parameters such as borer area and insect cocoon length) to correlate with the age of pests, so as to quantify the correspondence between the feeding characteristics of pests of different ages and the leaf damage morphology.

[0033] In specific experiments, images of leaves infested by target pests (such as rice leaf roller and diamondback moth) were collected, covering instars 1–6, with ≥50 leaf images per instar group. The following key characteristics were recorded: borer morphology (size, shape, edge features), frass state (granular or powdery, moisture level), and burrow structure (length, opening method, density of silk threads). The instar criteria for rice leaf roller can be set as follows: 1st instar presents as pinhead-sized translucent white spots without burrows; 2nd instar presents burrows 2–4.5 cm long with moist white streaks on the leaves; 5th instar presents burrows scorched white and cracked, with dry powdery frass.

[0034] For example, images of leaves infested by pests of different instars (such as fall armyworm larvae) are collected using a fixed 8-megapixel camera (30–50 cm from the leaf, with constant light source). These images cover instars 1–6, with ≥200 images per instar. The damaged areas (such as boreholes and gnawing edges) and corresponding instar labels are marked. The images are then flipped, rotated, and Gaussian noise is added. Mixup and Mosaic enhancement are used. An improved UNet++ (such as MRES-UNet++) is used to segment the damaged areas, and the insect length, head width, and borehole area are calculated. Based on a YOLOv5 or YOLOv11 model, the segmented damaged feature map is input, and the instar classification is output. Precision, recall, and F1 score (target F1 ≥ 0.86 when the confidence threshold is 0.475) are used as evaluation metrics for the model.

[0035] Please refer to Table 1, which provides the morphological criteria for fall armyworm larvae.

[0036] Table 1

[0037] S150, based on the social structure of each type of pest, determine the pesticide resistance corresponding to that pest from a pre-set table of multiple pesticide resistance structures.

[0038] Because older insects have higher metabolic enzyme activity and thicker epidermis, it can be concluded that the higher the proportion of older insect pests, the stronger the population's resistance to pesticides.

[0039] In experiments constructing multiple antibiotic resistance structure control tables, control and experimental groups can be set up. The control group is a natural population evenly distributed across all ages. The experimental groups can be set as follows: an older dominant group, with 5–6-year-old individuals accounting for ≥70%; and a younger dominant group, with 1–2-year-old individuals accounting for ≥70%. Antibiotic resistance is tested on the samples. At the gene level, PCR is used to detect the expression level of the CYP450 enzyme gene (the core gene of antibiotic resistance) in the older group. Antibiotic resistance can also be detected at the phenotypic level, and the median lethal concentration (LD50) can be calculated. Resistance ratio (RR) = experimental group / Control group High resistance is defined as RR > 106.

[0040] Then, a correlation analysis was conducted between the proportion of older pests and the population's resistance to pesticides: the correlation between the proportion of older larvae and the RR value was statistically analyzed (expected positive correlation, r≥0.6). The following correlation was found between age structure and pesticide resistance: the thickened epidermis of older larvae hinders pesticide penetration (such as the waxy layer of scale insects), and the expression level of CYP450 enzyme in 5-6 instar larvae can reach 3-5 times that of younger larvae, which indicates the enhanced activity of detoxification enzyme.

[0041] Therefore, a higher proportion of older pests indicates stronger pesticide resistance in a population. The specific correlation can be further quantified through experiments to obtain a pesticide resistance structure comparison table.

[0042] The drug resistance structure comparison table includes the relationship between drug resistance multiples and the social structure of pests. As the drug resistance multiples in the drug resistance structure comparison table increase from low to high, the difference between adjacent drug resistance multiples increases.

[0043] S160 predicts future insect infestation data for each target field based on current insect infestation data and historical insect infestation data for multiple target fields.

[0044] Based on the migratory nature of pests, early prevention and control measures can be implemented to further improve pest control effectiveness. Specifically, S160 includes S161 to S164.

[0045] S161. Based on the current insect pest data and the historical insect pest data of multiple target fields, construct a random variable X = (X1, X2, ..., Xn), where Xi is a three-dimensional matrix, the data at time i, i and n are both positive integers, and i ≤ n; S162, According to the principal component analysis method, extract the feature vector Zi for each Xi in the random variable X to obtain the feature variable Z = (Z1, Z2, ..., Zn); S163, Perform vector autoregression analysis on the feature variable Z to obtain the eigenvector Zn+1 corresponding to the predicted time n+1.

[0046] S164. Map the feature vector Zn+1 back to the three-dimensional matrix to obtain the three-dimensional matrix Xn+1 corresponding to time n+1. Xn+1 is used to represent the insect situation data corresponding to time n+1.

[0047] Preferably, two dimensions of the three-dimensional matrix Xi are used to represent the positional coordinates between each target field area, one dimension of the three-dimensional matrix Xi is used to represent the pest category, and the element xpjk in the three-dimensional matrix Xi is used to represent the quantity data of the k-th type of pest at coordinates (p, j).

[0048] The quantity data of the k-th type of pest in the element xpjk of the three-dimensional matrix Xi is determined based on the quantity and developmental stage of the k-th type of pest at the (p,j) coordinate. Specifically, the quantity coefficient of the k-th type of pest at the (p,j) coordinate is determined based on the proportion of each developmental stage in the social structure of the k-th type of pest at the (p,j) coordinate. The quantity data of the k-th type of pest at the (p,j) coordinate is determined based on the quantity and data coefficient of the k-th type of pest at the (p,j) coordinate. That is, different weights are assigned to pests according to their migratory ability at different ages to improve the accuracy of prediction. The specific weights can be set according to the actual situation.

[0049] S170 determines the pesticide ratio and pesticide-water ratio for irrigation of each target field based on the pesticide resistance of each type of pest, the crop growth stage, and future pest data for each target field.

[0050] Specifically, the effectiveness of the active ingredients in the pesticide is determined based on the resistance of each type of pest, and the pesticide ratio and pesticide-water ratio are determined for irrigation of each target field based on the effectiveness of each active ingredient.

[0051] This application embodiment determines the current pest situation data of each target field area through crop images, determines the social structure of each type of pest based on the pest situation data, and determines the corresponding pesticide resistance of each pest from multiple preset pesticide resistance structure comparison tables based on the social structure of each type of pest. The social structure of the pest can ignore the influence of pest migration to truly reflect the pesticide resistance of the pest, thereby more accurately determining the pesticide resistance of the pest and more effectively defending against pests.

[0052] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0053] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A water-saving agricultural irrigation method, characterized in that, The method includes: Acquire multiple crop images from multiple target field areas; Determine the crop growth stage in the corresponding target field area based on the crop image; Pest identification is performed based on the crop images to determine the current pest situation data for each target field area; Determine the social structure of each type of pest based on the current pest situation data; Based on the social structure of each type of pest, the corresponding pesticide resistance is determined from a pre-set pesticide resistance structure comparison table. The pesticide formulation and pesticide-to-water ratio for each target field area are determined based on the pesticide resistance of each type of pest and the growth stage of the crop. Before determining the pesticide formulation and pesticide-to-water ratio for irrigation of each target field based on the pesticide resistance of each type of pest and the crop growth stage, the method further includes: Predict future insect infestation data for each target field area based on the current insect infestation data and the historical insect infestation data of the multiple target fields; The determination of pesticide formulation and pesticide-to-water ratio for irrigation of each target field area based on the pesticide resistance of each type of pest and the crop growth stage includes: The pesticide formulation and pesticide-water ratio for irrigation of each target field area are determined based on the pesticide resistance of each type of pest, the growth stage of the crop, and the future pest situation data for each target field area.

2. The method as described in claim 1, characterized in that, The step of predicting future insect infestation data for each target field based on the current insect infestation data and the historical insect infestation data of the multiple target fields includes: Based on the current insect pest data and the historical insect pest data of the multiple target fields, construct a random variable X = (X1, X2, ..., Xn), where Xi is a three-dimensional matrix, the data at time i, i and n are both positive integers, and i ≤ n; For each Xi in the random variable X, extract the feature vector Zi to obtain the feature variable Z = (Z1, Z2, ..., Zn). Vector autoregression analysis is performed on the feature variable Z to obtain the feature vector Zn+1 corresponding to the predicted time n+1. The feature vector Zn+1 is mapped back to a three-dimensional matrix to obtain the three-dimensional matrix Xn+1 corresponding to time n+1. Xn+1 is used to characterize the insect situation data corresponding to time n+1.

3. The method as described in claim 2, characterized in that, The step of extracting a feature vector Zi for each Xi in the random variable X to obtain the feature variable Z = (Z1, Z2, ..., Zn) includes: Based on the principal component analysis method, the feature vector Zi is extracted from each Xi in the random variable X to obtain the feature variable Z = (Z1, Z2, ..., Zn).

4. The method as described in claim 3, characterized in that, Two dimensions of the three-dimensional matrix Xi are used to represent the positional coordinates between each target field area, one dimension of the three-dimensional matrix Xi is used to represent the pest category, and the element xpjk in the three-dimensional matrix Xi is used to represent the quantity data of the k-th type of pest at coordinates (p, j).

5. The method as described in claim 4, characterized in that, The pest data includes pest type, quantity, and developmental stage; the quantity data of the k-th type of pest in the element xpjk is determined based on the quantity and developmental stage of the k-th type of pest at the (p, j) coordinate.

6. The method as described in claim 5, characterized in that, The method further includes: The quantity coefficient of the k-th type of pest in the (p,j) coordinate is determined based on the proportion of each developmental stage in the social structure of the k-th type of pest in the (p,j) coordinate. The quantity data of the k-th type of pest in the (p, j) coordinate system is determined based on the quantity and data coefficient of the k-th type of pest in the (p, j) coordinate system.

7. The method as described in claim 6, characterized in that, The drug resistance structure comparison table includes a comparison relationship between drug resistance multiples and the social structure of pests, and as the drug resistance multiples in the drug resistance structure comparison table increase from low to high, the difference between adjacent drug resistance multiples increases.

8. The method as described in claim 7, characterized in that, The process of determining the pesticide formulation and pesticide-to-water ratio for irrigation of each target field area based on the pesticide resistance of each type of pest and the crop growth stage includes: The effectiveness of the active ingredient in the pesticide is determined based on the resistance of each type of pest. The pesticide formulation and pesticide-to-water ratio for each target field area should be determined based on the effectiveness of each active ingredient.

Images