A farmland irrigation control system

Abstract

This invention discloses a farmland irrigation control system, comprising: a data acquisition module for recording crop images of a target crop, importing them into a database, and acquiring the humidity threshold of the target crop and soil data of the target soil; an analysis module for extracting first shape features and color features from the crop images, analyzing the first shape features and color features to obtain the root and stem depth of the target crop, and analyzing the soil data to obtain the threshold level of the target soil; and a decision module for analyzing the root and stem depth to obtain the humidity monitoring range, detecting the humidity data within the humidity monitoring range by the data acquisition module, adjusting the humidity threshold based on the threshold level, generating a first difference, and making irrigation decisions. This invention, by employing a dual analysis strategy of shape and color features, accurately identifies the growth cycle, completely solving the problem of inaccurate irrigation depth caused by misjudgment of growth stages in traditional methods, and improving the matching accuracy of irrigation depth.

Description

Technical Field

[0001] This invention relates to the technical field of smart agricultural irrigation control, and more particularly to a farmland irrigation control system. Background Technology

[0002] In recent years, more and more farmland has begun to introduce mechanized irrigation equipment, and irrigation has become the first choice for many agricultural greenhouses, fruit trees and cotton industries. Therefore, how to achieve automated control of irrigation equipment has become the core issue of agricultural intelligence and informatization.

[0003] The introduction of intelligent irrigation systems in existing technologies has solved the drawbacks and inconveniences of manually switching ball valves, allowing for more precise and rational control of irrigation time and water volume. Simultaneously, parallel fertilization devices can also achieve automated fertilization. However, current intelligent irrigation systems lack a dynamic adjustment mechanism when collecting data, failing to adjust thresholds according to different crop growth stages, thus leading to resource waste. Summary of the Invention

[0004] The technical problem solved by this invention is to address the issue that traditional irrigation systems fail to accurately detect the actual moisture content of the crop's water absorption layer due to their inability to differentiate between different soil depths at different growth stages, leading to low irrigation efficiency and resource waste.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A farmland irrigation control system, comprising:

[0007] The acquisition module records image data of the target crop, preprocesses the image data, imports it into a database, and obtains the humidity threshold of the target crop and soil data of the target soil; the image data includes crop images;

[0008] The analysis module extracts the first shape and color features of the crop image, analyzes the first shape and color features to obtain the root and stem depth of the target crop, and analyzes the soil data to obtain the threshold level of the target soil.

[0009] The decision module is used to analyze the root depth to determine the humidity monitoring range;

[0010] The decision module controls the acquisition module to detect humidity data within the humidity monitoring range, adjusts the humidity threshold based on the threshold level, generates the first difference, and makes an irrigation decision to determine whether to irrigate.

[0011] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0012] The acquisition module records image data of the target crop in a first time period as the monitoring cycle, and preprocesses the image data, including noise reduction and filtering. The image data includes the crop shape and crop color.

[0013] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0014] The analysis module extracts the first shape features of the crop image after preprocessing, performs a first analysis on the first shape features to obtain the growth cycle of the target crop, performs a second analysis on the color features of the crop image based on the growth cycle to obtain the cycle progress of the target crop, and determines the root depth based on a pre-established mapping table of growth cycle, cycle progress and root depth.

[0015] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0016] The analysis logic for the first shape feature includes:

[0017] Access a crop database, which includes images of standard crops at different growth stages;

[0018] Obtain any standard crop image from standard crop images at different growth stages, extract the second shape feature of the any standard crop image based on the crop shape, calculate the first similarity between the first shape feature and the second shape feature using the cosine similarity formula, sort the first similarity in descending order, select the standard crop image corresponding to the first similarity with the largest value, record it as the first image, and identify the growth stage to which the first image belongs.

[0019] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0020] The analysis logic for the color features includes:

[0021] Access a crop database, which includes the color depth range of crops at each growth stage;

[0022] The crop image is converted into a grayscale image, and the grayscale value of each pixel in the grayscale image is obtained. Using the interval pixel sampling method, the grayscale difference between itself and its adjacent pixels is calculated with each interval pixel as the center. The adjacent pixels with the largest grayscale difference and the center pixel are stored in a temporary data array, and compared with a preset segmentation threshold. Grayscale values ​​with a maximum grayscale difference less than the preset segmentation threshold are obtained. The grayscale values ​​are grouped according to their values, and a corresponding number of storage arrays are set. The screening condition for the storage arrays is that the grayscale value falls within the grayscale value range of the corresponding group. When the maximum grayscale difference is greater than the preset segmentation threshold, the coordinate information of the adjacent pixels with the maximum grayscale difference and the center pixel are recorded into the corresponding storage arrays according to their corresponding grayscale values. Based on the coordinate information, the proportion of the number of pixels in each storage array in the grayscale image is calculated, and the storage arrays are sorted in descending order according to their proportions. The storage array with the largest proportion is recorded as the crop grayscale value array.

[0023] The color depth range of the crop corresponding to the growth cycle is converted into a standard grayscale value range. Based on the position information of the most concentrated grayscale value range in the crop grayscale value array within the standard grayscale value range, the corresponding progress percentage is generated, thereby determining the cycle progress of the target crop.

[0024] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0025] The analysis logic for the threshold level includes:

[0026] A soil database is introduced, which includes the water retention rate corresponding to each soil type;

[0027] Based on the soil data, obtain the soil type of the target soil, and based on the soil type, obtain the corresponding water retention rate;

[0028] Based on the different water retention rates, threshold levels are generated for each soil category.

[0029] The water retention rates are sorted in descending order, and the first, second, and third thresholds are set sequentially. Based on the threshold levels corresponding to each soil type, the corresponding humidity threshold is selected from the first, second, and third thresholds.

[0030] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0031] It also includes a prediction module:

[0032] The prediction module connects to the meteorological data system and acquires local weather data in a second time period, including rainfall data and wind data.

[0033] Based on wind data, the values ​​of the first, second, and third thresholds within the threshold level are adjusted to achieve dynamic adjustment of the humidity threshold; and based on rainfall data and the first difference, a water accumulation risk value is analyzed and generated, and a decision on whether to irrigate is made based on the water accumulation risk value.

[0034] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0035] The execution logic for dynamically adjusting the humidity threshold includes:

[0036] The wind data includes wind speed levels. The wind speed levels are compared with preset level thresholds. When the wind speed level exceeds the preset level threshold, the values ​​of the first threshold, the second threshold, and the third threshold are reduced by a preset ratio.

[0037] As a preferred embodiment of the farmland irrigation control system described in this invention, wherein:

[0038] The analysis logic for the water accumulation risk value includes:

[0039] Obtain the rainfall probability and rainfall level from the rainfall data;

[0040] Calculate the product of the rainfall probability and the first difference, and compare it with a preset rainfall threshold. If the product of the rainfall probability and the first difference exceeds the preset rainfall threshold, make an irrigation decision to determine whether to irrigate; if the product of the rainfall probability and the first difference is less than the preset rainfall threshold, generate a pause irrigation command.

[0041] A method for controlling farmland irrigation, applied to the aforementioned farmland irrigation control system, includes the following steps:

[0042] Step S100: The acquisition module records the image data of the target crop and obtains the humidity threshold of the target crop and the soil data of the target soil.

[0043] In step S200, the analysis module analyzes the first shape feature and color feature to obtain the root and stem depth of the target crop, and analyzes the soil data to obtain the threshold level of the target soil.

[0044] In step S300, the decision module obtains the humidity monitoring range based on the root depth, controls the acquisition module to detect the humidity data within the humidity monitoring range, adjusts the humidity threshold in combination with the threshold level, generates the first difference, and makes an irrigation decision to determine whether to irrigate.

[0045] The beneficial effects of this invention are:

[0046] First, the farmland irrigation control system provided by this invention accurately identifies the growth cycle by adopting a dual analysis strategy of shape and color features, which completely solves the problem of inaccurate irrigation depth caused by misjudgment of growth stage in traditional methods, and improves the matching degree of irrigation depth.

[0047] Secondly, the farmland irrigation control system provided by this invention establishes a dynamic grading model of soil water retention rate and three threshold levels (first threshold, second threshold, and third threshold), and automatically matches the threshold level according to different soil types (such as sandy soil with low water retention and clay soil with high water retention), which can effectively avoid root hypoxia damage.

[0048] Third, the farmland irrigation control system provided by the present invention uses a wind speed-threshold linkage mechanism (automatically reducing the threshold at each level to reduce evaporation when the wind speed exceeds the preset level) and combines the rainfall probability and humidity difference (first difference) to construct an overflow risk quantification model, thereby reducing over-irrigation events caused by rainstorms. Attached Figure Description

[0049] Figure 1 This is a schematic flowchart of a farmland irrigation control system provided in one embodiment of the present invention. Detailed Implementation

[0050] 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.

[0051] In recent years, more and more farmland has begun to introduce mechanized irrigation equipment, and irrigation has become the first choice for many agricultural greenhouses, fruit trees and cotton industries. Therefore, how to achieve automated control of irrigation equipment has become the core issue of agricultural intelligence and informatization.

[0052] The introduction of intelligent irrigation systems in existing technologies has solved the drawbacks and inconveniences of manually switching ball valves, allowing for more precise and rational control of irrigation time and water volume. Simultaneously, parallel fertilization devices can also achieve automated fertilization. However, current intelligent irrigation systems lack a dynamic adjustment mechanism when collecting data, failing to adjust thresholds according to different crop growth stages, thus leading to resource waste.

[0053] Example 1, refer to Figure 1This invention provides a farmland irrigation control system, comprising: a data acquisition module for recording image data of a target crop, preprocessing the image data, importing it into a database, and acquiring the humidity threshold of the target crop and soil data of the target soil; the image data includes crop images; an analysis module for extracting first shape features and color features from the crop images, analyzing the first shape features and color features to obtain the root and stem depth of the target crop, and analyzing the soil data to obtain the threshold level of the target soil; a decision module for analyzing the root and stem depth to obtain the humidity monitoring range; the decision module controls the data acquisition module to detect humidity data within the humidity monitoring range, adjusts the humidity threshold based on the threshold level, generates a first difference, makes an irrigation decision, and determines whether to irrigate.

[0054] The aforementioned farmland irrigation control system includes the following execution steps:

[0055] In step S100, the acquisition module records the image data of the target crop and obtains the humidity threshold of the target crop and the soil data of the target soil.

[0056] In step S110, the acquisition module records image data of the target crop using a first time period as the monitoring cycle. The image data is preprocessed, including noise reduction and filtering. The image data includes the crop's shape and color. The first time period is measured in months, typically one month.

[0057] In step S200, the analysis module analyzes the first shape and color features to obtain the root depth of the target crop and analyzes the soil data to obtain the threshold level of the target soil. By obtaining the root depth of the target crop, the humidity of the current root area of ​​the target crop can be monitored to ensure that the target crop can be irrigated appropriately. Due to the thickness of the root walls, the bottom roots of the crop have the thinnest walls and are the main areas for water absorption.

[0058] Step S210: The analysis module extracts the first shape features of the crop image after preprocessing, performs a first analysis on the first shape features to obtain the growth cycle of the target crop, performs a second analysis on the color features of the crop image based on the growth cycle to obtain the cycle progress of the target crop, and determines the root depth based on the pre-established mapping relationship table between the growth cycle, cycle progress and root depth.

[0059] Step S211: A crop database is accessed, which includes standard crop images at different growth stages. Any standard crop image is obtained from these images. A second shape feature of the standard crop is extracted based on its shape. The first similarity between the first and second shape features is calculated using the cosine similarity formula. The first similarities are then sorted in descending order. The standard crop image with the highest first similarity score is selected and designated as the first image. The growth stage of the first image is then identified. The first and second shape features are extracted using machine vision. This allows for automatic determination of the current target crop's growth state through machine vision. Different growth stages of plants exhibit different shape features, which can be automatically identified using sophisticated machine vision technology.

[0060] Step S212: Call the crop database, which includes the color depth range of crops at each growth stage;

[0061] The crop image is converted to a grayscale image, and the grayscale value of each pixel within the grayscale image is obtained. An interval pixel sampling method is used, with each interval pixel as the center, such as sampling the grayscale values ​​of pixels at coordinates (1,1), (1,3), (1,n), (3,1), and (3,3). The grayscale difference between each pixel and its adjacent pixels is then calculated. The adjacent pixels with the largest grayscale difference and the center pixel are stored in a temporary data array, which is then compared with a preset segmentation threshold. Grayscale values ​​with a maximum grayscale difference less than the preset threshold are obtained. Based on the grayscale values... The values ​​are grouped into clusters, and a corresponding number of storage arrays are set. The screening condition for the storage arrays is that the gray values ​​fall within the gray value range of the corresponding cluster. When the maximum gray value difference is greater than the preset segmentation threshold, the coordinate information of the adjacent pixels and the center pixel of the maximum gray value difference are recorded into the corresponding storage arrays according to their corresponding gray values. Based on the coordinate information, the proportion of the number of pixels in the gray image in each storage array is calculated, and the storage arrays are sorted in descending order according to the proportion. The storage array with the largest proportion is recorded as the crop gray value array.

[0062] The color depth range of the crop corresponding to the growth cycle is converted into a standard grayscale value range. Based on the position information of the most concentrated grayscale value range in the crop grayscale value array within the standard grayscale value range, the corresponding progress percentage is generated, thereby determining the cycle progress of the target crop.

[0063] Therefore, this application can automatically determine the approximate stage of the current growth state by recognizing color features, thereby improving the recognition effect and efficiency. For example, at the end of the germination period, the color of the crop becomes significantly darker, maintaining a dark green color, which is clearly different from the light green color of crops that have just entered the germination period.

[0064] Step S220: A soil database is introduced, including the water retention rate corresponding to each soil type. The soil type of the target soil is obtained based on the soil data, and the corresponding water retention rate is obtained based on the soil type. Threshold levels are generated for each soil type according to the different water retention rates. The water retention rates are sorted in descending order, and the first, second, and third thresholds are set sequentially. A corresponding humidity threshold is selected from the first, second, and third thresholds based on the threshold levels corresponding to each soil type. Furthermore, based on the differences in water retention rates between different soils, sandy soil and black soil are significantly different. Therefore, this application can automatically match threshold levels according to different soil types (such as sandy soil with low water retention and clay soil with high water retention), effectively preventing excessive soil moisture content and crop root hypoxia damage due to over-irrigation.

[0065] In step S300, the decision module obtains the humidity monitoring range based on the root depth, controls the acquisition module to detect the humidity data within the humidity monitoring range, adjusts the humidity threshold in combination with the threshold level, generates the first difference, and makes an irrigation decision to determine whether to irrigate.

[0066] In step S400, the prediction module connects to the meteorological data system and acquires local weather data, including rainfall and wind data, using a second time period as the cycle. Based on the wind data, the values ​​of the first, second, and third thresholds within the threshold level are adjusted to dynamically adjust the humidity threshold. Furthermore, based on the rainfall data and the first difference, a waterlogging risk value is analyzed and generated. Based on this waterlogging risk value, a decision is made regarding whether to irrigate. The second time period is set in weeks, according to the frequency of weather changes, typically one week.

[0067] Step S410: Wind data includes wind speed level. The wind speed level is compared with a preset level threshold. When the wind speed level exceeds the preset level threshold, the values ​​of the first threshold, second threshold, and third threshold are reduced by a preset ratio. Therefore, when the wind speed is high, the rate of soil moisture evaporation increases, effectively preventing soil water shortage during the automatic check cycle due to greater fluctuations in soil moisture caused by air movement.

[0068] Step S420: Obtain the rainfall probability and rainfall level from the rainfall data; calculate the product of the rainfall probability and the first difference, and compare it with a preset rainfall threshold. When the product of the rainfall probability and the first difference exceeds the preset rainfall threshold, make an irrigation decision to determine whether to irrigate; when the product of the rainfall probability and the first difference is less than the preset rainfall threshold, generate a pause irrigation command. In this way, this application can effectively avoid the problem of over-irrigation of crops due to rain after irrigation.

[0069] Example 2: This application provides another method for identifying the crop cycle progress, which includes the following steps:

[0070] Step a, color feature extraction:

[0071] First, a color space conversion is performed, transforming the preprocessed image from the RGB color space to a color space more suitable for color analysis, such as HSV or LAB.

[0072] Next, color features are extracted, specifically the color features of the region of interest (ROI) in the image, such as calculating the average hue, saturation, and brightness of the ROI.

[0073] Step b, compare with the standard image:

[0074] First, load standard images. These are predefined standard images that represent the color characteristics of crops at different maturity stages.

[0075] Next, the color features of the standard image are extracted, and the same preprocessing and color feature extraction operations are performed on the standard image.

[0076] Finally, the color feature differences are calculated, specifically the differences between the color features of the real-time image and the color features of the standard image.

[0077] Step c, determine maturity:

[0078] Based on the magnitude of the color feature differences, determine which standard image the real-time image is closest to (standard images are collected at the beginning and end of the cycle, and maturity is judged using the stage where the real-time image is in the middle color feature range), thereby determining the maturity stage of the crop.

[0079] 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.

[0080] 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 farmland irrigation control system, characterized in that, include: The acquisition module records image data of the target crop, preprocesses the image data, imports it into a database, and obtains the humidity threshold of the target crop and soil data of the target soil; the image data includes crop images; The analysis module extracts the first shape and color features of the crop image, analyzes the first shape and color features to obtain the root and stem depth of the target crop, and analyzes the soil data to obtain the threshold level of the target soil. The decision module is used to analyze the root depth to determine the humidity monitoring range; The decision module controls the acquisition module to detect humidity data within the humidity monitoring range, adjusts the humidity threshold based on the threshold level, generates the first difference, and makes an irrigation decision to determine whether to irrigate. The analysis logic for the threshold level includes: A soil database is introduced, which includes the water retention rate corresponding to each soil type; Based on the soil data, obtain the soil type of the target soil, and based on the soil type, obtain the corresponding water retention rate; Based on the different water retention rates, threshold levels are generated for each soil category. The water retention rates are sorted in descending order, and the first, second, and third thresholds are set sequentially. Based on the threshold levels corresponding to each soil type, the corresponding humidity threshold is selected from the first, second, and third thresholds. It also includes a prediction module: The prediction module connects to the meteorological data system and acquires local weather data in a second time period, including rainfall data and wind data. Based on wind data, the values ​​of the first, second, and third thresholds within the threshold level are adjusted to achieve dynamic adjustment of the humidity threshold; and based on rainfall data and the first difference, a water accumulation risk value is analyzed and generated, and a decision on whether to irrigate is made based on the water accumulation risk value. The execution logic for dynamically adjusting the humidity threshold includes: Wind data includes wind speed levels. The wind speed levels are compared with preset level thresholds. When the wind speed level exceeds the preset level threshold, the values ​​of the first threshold, the second threshold, and the third threshold are reduced by a preset ratio. The analysis logic for the water accumulation risk value includes: Obtain the rainfall probability and rainfall level from the rainfall data; Calculate the product of the rainfall probability and the first difference, and compare it with a preset rainfall threshold. If the product of the rainfall probability and the first difference exceeds the preset rainfall threshold, make an irrigation decision to determine whether to irrigate; if the product of the rainfall probability and the first difference is less than the preset rainfall threshold, generate a pause irrigation command.

2. The farmland irrigation control system as described in claim 1, characterized in that: The acquisition module records image data of the target crop in a first time period as the monitoring cycle, and preprocesses the image data, including noise reduction and filtering. The image data includes the crop shape and crop color.

3. The farmland irrigation control system as described in claim 1, characterized in that: The analysis module extracts the first shape features of the crop image after preprocessing, performs a first analysis on the first shape features to obtain the growth cycle of the target crop, performs a second analysis on the color features of the crop image based on the growth cycle to obtain the cycle progress of the target crop, and determines the root depth based on a pre-established mapping table of growth cycle, cycle progress and root depth.

4. The farmland irrigation control system as described in claim 3, characterized in that, The analysis logic for the first shape feature includes: Access a crop database, which includes images of standard crops at different growth stages; Obtain any standard crop image from standard crop images at different growth stages, extract the second shape feature of the any standard crop image based on the crop shape, calculate the first similarity between the first shape feature and the second shape feature using the cosine similarity formula, sort the first similarity in descending order, select the standard crop image corresponding to the first similarity with the largest value, record it as the first image, and identify the growth stage to which the first image belongs.

5. The farmland irrigation control system as described in claim 3, characterized in that, The analysis logic for the color features includes: Access a crop database, which includes the color depth range for each growth stage of the crop; The crop image is converted into a grayscale image, and the grayscale value of each pixel in the grayscale image is obtained. Using the interval pixel sampling method, the grayscale difference between itself and its adjacent pixels is calculated with each interval pixel as the center. The adjacent pixels with the largest grayscale difference and the center pixel are stored in a temporary data array, and compared with a preset segmentation threshold. Grayscale values ​​with a maximum grayscale difference less than the preset segmentation threshold are obtained. The grayscale values ​​are grouped according to their values, and a corresponding number of storage arrays are set. The screening condition for the storage arrays is that the grayscale value falls within the grayscale value range of the corresponding group. When the maximum grayscale difference is greater than the preset segmentation threshold, the coordinate information of the adjacent pixels with the maximum grayscale difference and the center pixel are recorded into the corresponding storage arrays according to their corresponding grayscale values. Based on the coordinate information, the proportion of the number of pixels in each storage array in the grayscale image is calculated, and the storage arrays are sorted in descending order according to their proportions. The storage array with the largest proportion is recorded as the crop grayscale value array. The color depth range of the crop corresponding to the growth cycle is converted into a standard grayscale value range. Based on the position information of the most concentrated grayscale value range in the crop grayscale value array within the standard grayscale value range, the corresponding progress percentage is generated, thereby determining the cycle progress of the target crop.

6. A method for controlling farmland irrigation, applied to the farmland irrigation control system according to any one of claims 1 to 5, characterized in that, Includes the following steps: Step S100: The acquisition module records the image data of the target crop and obtains the humidity threshold of the target crop and the soil data of the target soil. In step S200, the analysis module analyzes the first shape feature and color feature to obtain the root and stem depth of the target crop, and analyzes the soil data to obtain the threshold level of the target soil. In step S300, the decision module obtains the humidity monitoring range based on the root depth, controls the acquisition module to detect the humidity data within the humidity monitoring range, adjusts the humidity threshold in combination with the threshold level, generates the first difference, and makes an irrigation decision to determine whether to irrigate.

Images