Greenhouse automatic spraying control system

Abstract

The invention discloses an automatic spraying control system for a greenhouse, which relates to the technical field of intelligent agriculture and comprises an information acquisition module, a CMOS (complementary metal oxide semiconductor) camera, an image processing system, a crop image state analysis module, an intelligent decision and spraying control module and spraying equipment. Through the greenhouse automatic spraying control system, environment information in a greenhouse is obtained in real time, images of crops are collected, the images are preprocessed, the health condition and the moisture stress degree of the crops are analyzed, the most appropriate spraying strategy is made, and spraying equipment executes the spraying strategy and conducts precise spraying on the crops. In addition, by adding a small lifting device and introducing a learning model, the spraying effect is further optimized, the utilization efficiency of water resources is improved, the complexity and labor intensity of manual operation are reduced, and the scientificity and efficiency of agricultural production are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent agriculture, and in particular to an automatic sprinkler control system for a greenhouse. Background Art

[0002] In agricultural production, especially in the production of greenhouse vegetables, flowers, and fruit trees, in order to promote the growth of crops, it is necessary to control the environmental information in the greenhouse, such as air temperature and humidity, within a reasonable range. The method of controlling the environmental information in the greenhouse is usually mainly spraying water, and the spraying operation depends on corresponding equipment, such as spraying equipment consisting of water pumps and nozzles. However, traditional spraying methods often rely on experience judgment, or manual observation of the soil moisture on the spot to determine whether spraying is needed. There is a lack of objective and accurate measurement data, and it is impossible to accurately control the amount of water sprayed according to the actual environmental conditions. In addition, traditional spraying methods often use large-flow, large-area spraying methods, which can easily lead to a large consumption of water resources; inappropriate spraying intensity may cause soil erosion, destroy the soil structure, and affect the soil fertility and water retention capacity. Summary of the invention

[0003] The present application solves the problems of uneven spraying, waste of resources, poor crop growth environment and cumbersome manual operation in the prior art by providing an automatic greenhouse spray control system, realizes intelligent and precise spray control, and improves the scientificity and efficiency of agricultural production.

[0004] The present application provides a greenhouse automatic sprinkler control system, comprising: Step 1: The information collection module collects the indoor environmental information of the greenhouse, and collects the carbon dioxide concentration, air temperature and humidity, soil humidity, and light intensity in the greenhouse through the carbon dioxide concentration sensor, air temperature and humidity sensor, soil humidity sensor, and light sensor; Step 2, collect crop images through CMOS camera to capture their subtle features; Step 3, transmitting the crop image to an image processing system, performing image preprocessing on the crop image by grayscale processing, noise removal, and image feature enhancement to obtain a preprocessed crop image; Step 4, analyzing the state of the preprocessed crop images to obtain the health state of the crops or whether they are under water stress; Step 5: When formulating a spraying strategy, in addition to considering the health status, water stress level and environmental information of the crops, the height data of the crops also needs to be considered; the camera is used to obtain the image of the crop height measuring scale, and the height data of the crops is obtained in real time through the combination of image processing and computer vision algorithms, and transmitted to the control system in the small lifting device. The control system adjusts the position of the small lifting device according to the height data to ensure that the water flow of the spraying equipment can cover all crops; Step 6, the intelligent decision-making and spraying control module conducts a comprehensive analysis based on the health status of the crops and the degree of water stress, combined with the environmental information and height data of the environment in which the crops are located, and formulates the most appropriate spraying strategy, which can be suitable for crops with different growth conditions; Step 7: The spraying equipment executes the spraying strategy to accurately spray the crops in the greenhouse to meet the moisture needs of the crops.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages: Through the information acquisition module and CMOS camera, the indoor environment information and crop images of the greenhouse are collected in real time, providing a reliable data basis for precise spraying; through image processing and computer vision algorithms, the crop images are preprocessed and the state analysis is performed to accurately judge the health status and degree of water stress of the crops, providing a basis for formulating personalized spraying strategies; by introducing a small lifting device, the position of the spraying equipment is adjusted according to the height of the crops, thus achieving comprehensive and uniform spraying and solving the problem of uneven spraying. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 It is a schematic diagram of a flow chart of an automatic sprinkler control system for a greenhouse in an embodiment of the present invention; Figure 2 A schematic diagram of the process of designing and applying a small lifting device in an embodiment of the present invention; Figure 3 A schematic diagram of the process of building a learning model and application in an embodiment of the present invention; Figure 4 This is a framework diagram of an automatic sprinkler control system for a greenhouse in an embodiment of the present invention; Figure 5 It is a framework diagram of a small lifting device in an embodiment of the present invention; DETAILED DESCRIPTION

[0007] To facilitate the understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0008] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which the present invention belongs; the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more related listed items.

[0009] In view of the uneven spraying effect and difficulty in adapting to changes in crop growth in the greenhouse spraying system in the existing technology, this application adopts a variety of sensor devices, CMOS cameras, advanced image processing and analysis algorithms, and machine learning models to achieve intelligent and precise greenhouse spraying.

[0010] Embodiment 1: Figure 1 As shown, a greenhouse automatic sprinkler control system includes the following steps: Step 1, collect the indoor environmental information of the greenhouse through the information collection module: use a carbon dioxide concentration sensor with an accuracy of less than ±10ppm or even higher, an air temperature and humidity sensor with an accuracy of ±0.1°C or higher, a soil moisture sensor with an accuracy of ±3% (m³ / m³) or higher in the range of 0-50% (m³ / m³), and a light sensor with an error of no more than 2LUX under 100LUX light, which can accurately and real-time capture various environmental information in the greenhouse.

[0011] Real-time monitoring of carbon dioxide concentration levels in the greenhouse through carbon dioxide concentration sensors is crucial to understanding the photosynthesis efficiency of plants and the air quality in the greenhouse; using air temperature and humidity sensors, accurately measuring the air temperature and humidity in the greenhouse, these two parameters are crucial to the growth and development of plants, too high or too low temperature and humidity may have an adverse effect on plants; deploying soil moisture sensors to penetrate deep into the soil and directly measure the moisture content of the soil, so as to determine the moisture absorption status of the plant roots; with the help of light sensors, accurately measuring the light intensity in the greenhouse, because light is the main source of energy for plants to photosynthesize and synthesize organic matter, and its suitability directly affects the growth rate and quality of plants.

[0012] Step 2: Take photos of crops in the greenhouse with a CMOS camera: In order to accurately grasp the growth status of various crops in the greenhouse, a CMOS camera is used to collect images of crops. In this process, CMOS cameras are placed in high places, corners, blind spots and areas with high planting density in the greenhouse to ensure that all crops can be fully covered; CMOS cameras can capture the subtle features of crop leaves, including their color, shape, and texture; high frame rate image acquisition improves the ability to monitor the growth status of crops, providing a rich and accurate information basis for subsequent image analysis and data processing.

[0013] Step 3: The collected crop images are transmitted to the image processing system, and the crop images are preprocessed by grayscale processing, noise removal, and image feature enhancement to obtain the preprocessed crop images: Step 301, grayscale processing: The conversion process performs weighted average on the values ​​of the three channels R, G, and B. The specific calculation formula is: grayscale value = 0.3R + 0.59G + 0.11B; the collected crop image is converted into a grayscale image, which reduces the color information of the image, but the key structural information and defect features can be retained in the grayscale image, thereby reducing the amount of data processing and improving the processing speed; Step 302, removing noise: Gaussian filtering and median filtering can make the image smooth while maintaining the edge information of the image; Gaussian filtering is a linear filtering technology, which removes noise by weighted averaging each pixel point in the crop image and the pixels in its neighborhood, so as to make the crop image smooth; median filtering is a nonlinear filtering technology, which selects a neighborhood size of 3*3 or 5*5, takes all pixel values ​​in the neighborhood of each pixel point in the crop image, sorts these pixels from small to large, and selects the median after sorting as the new value of the pixel point, so as to remove noise while maintaining the edge information of the crop image; Step 303: Enhance image features: Use edge detection operators to detect edges in crop images, and highlight edges by enhancing edge pixels; use texture filters to enhance texture information in crop images, and apply local contrast enhancement to improve texture recognizability; enhance high-frequency components of crop images through sharpening filters to improve clarity and detail of crop images; equalize the grayscale histogram of crop images to improve the contrast of crop images and make their features more prominent.

[0014] Step 4: Perform status analysis on the preprocessed crop images to determine the health status of the crops or whether they are under water stress: Step 401, in the grayscale image, the health status of the crop is comprehensively evaluated by the distribution of pixel values, the roughness of the texture, and the regularity of the shape. Specifically, the grayscale values ​​of healthy crop leaves are mainly concentrated between 100 and 150, while diseased or damaged leaves may have areas with abnormal grayscale values, which are specifically manifested as follows: the diseased area shows a significant increase or decrease in grayscale value, forming a bright or dark spot with a clear contrast with the surrounding healthy tissue; the disease causes necrosis, yellowing, spots and other symptoms in local areas of the leaves, and these areas appear as patches with abnormally reduced or increased grayscale values ​​in the grayscale image.

[0015] Gray-level co-occurrence matrix (GLCM) features are used to extract the contrast of the image. An image with high contrast indicates that its texture grooves are deep, which means that the leaf surface is rough or damaged by pests and diseases. Low contrast means that the texture is fine, the leaf surface is relatively smooth, and the leaf is relatively healthy. In addition, the regularity of the shape is also an important basis for evaluating the health of the leaves. Healthy leaves are usually symmetrical from left to right or from top to bottom, while sick or damaged leaves are the opposite.

[0016] Step 402, in the grayscale image, by observing whether the edges and overall state of the leaves of the crops appear distorted, it can be further determined whether the crops are under water stress; the edges of the leaves under water stress are no longer continuous or straight but are obviously bent or folded, and the grayscale value distribution of the edge area shows alternating light and dark stripes or block areas; in addition, the leaves under water stress often appear wilted as a whole, the color becomes darker, and the glossiness decreases.

[0017] The specific value can be adjusted dynamically according to the actual situation.

[0018] Step 5: The intelligent decision-making and spraying control module uses statistics and machine learning to analyze and formulate appropriate spraying strategies based on the health status of crops and the degree of water stress, combined with the specific environmental information of the crops, which can be suitable for crops in different regions and different growth conditions.

[0019] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application uses a variety of sensor devices, CMOS cameras, and advanced image processing and analysis algorithms to achieve intelligent and precise automatic spraying in greenhouses. Specifically, by collecting environmental information and crop images in the greenhouse in real time, preprocessing and analyzing the crop images, it is possible to accurately judge the health status and degree of water stress of crops, thereby formulating the most appropriate spraying strategy. This technical solution not only improves the scientificity and efficiency of agricultural production, realizes resource conservation, and helps achieve sustainable agricultural development.

[0020] Example 2: In Example 1, due to the differences in the height of crops, it is often difficult for the spraying equipment to fully cover all crops, resulting in uneven spraying effects. Specifically, when the growth cycles of crops are different or the types are diverse, their heights will also be different. Some taller crops may block the water flow of the spraying equipment, making it impossible for short crops to be fully sprayed, while too high a spray may result in a waste of water resources and low spraying efficiency. In addition, the growth density and leaf distribution of crops will also affect the uniformity of the spray, further exacerbating the problem of incomplete spray coverage.

[0021] Therefore, the embodiment of the present application is optimized based on the above embodiment, and specifically includes the following steps: Step 601, the structure of the small lifting device: The small lifting device is composed of a motor, a transmission system, a lifting platform and a control system. The motor provides power, the transmission system is responsible for converting the rotational motion of the motor into the vertical motion of the lifting platform, and the control system controls the operation of the motor according to the height data of the crops. The lifting platform can carry the spraying equipment up and down and adjust the position of the spraying equipment according to the actual height of the crops.

[0022] Step 602, adjust the position of the small lifting device: use the camera to obtain the image of the crop height measuring scale, and through the combination of image processing and computer vision algorithm, obtain the height data of the crop in real time and transmit it to the control system. The control system adjusts the position of the small lifting device according to the height data to ensure that the water flow of the sprinkler can cover all crops.

[0023] Step 603, workflow of the greenhouse automatic spray control system: First, the information acquisition module collects the greenhouse indoor environment information and crop height data. Then, the CMOS camera collects images of the crops, and performs image preprocessing and state analysis. Next, the intelligent decision-making and spray control module formulates a spray strategy based on the health status, water stress level, environmental information, growth stage and height data of the crops. Finally, the small lifting device adjusts the position of the spray equipment according to the spray strategy, and the spray equipment executes the spray strategy to achieve comprehensive and uniform spraying.

[0024] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application can adjust the position of the spraying equipment according to the height data of the crops by adding a small lifting device, ensuring that the water flow can cover all crops, solving the problem of uneven spraying, and improving the uniformity and accuracy of spraying. This technical solution not only improves the utilization efficiency of water resources, but also enhances the growth environment of crops, helps to improve their yield and quality, while also reducing the tediousness and labor intensity of manual operations, and improving the scientificity and efficiency of agricultural production.

[0025] Embodiment 3: Based on Embodiment 1 and Embodiment 2, this embodiment introduces a machine learning model to further optimize the performance and intelligence level of the greenhouse automatic sprinkler control system. Specifically, the following steps are included: Step 701, building a learning model: extracting features from the preprocessed crop images, including pixel value distribution, texture roughness, and shape regularity, to form feature vectors; building a training data set in combination with environmental information and crop growth conditions; selecting a random model machine learning algorithm, using the training data set to perform model training, and obtaining a model that can predict crop health and water stress levels; Step 702, model application and intelligent spraying: deploy the trained model to the greenhouse automatic spraying control system, and predict the collected leaf images and environmental information data in real time to obtain the health status and water stress level of the crops; formulate personalized spraying strategies based on the prediction results, combined with the specific environmental information and growth conditions of the crops. At the same time, consider the height data of the crops, use the CMOS camera to obtain crop images, and obtain height data in real time; according to the height data, control the small lifting device to adjust the position of the spraying equipment to ensure that the water flow can cover all crops and achieve comprehensive and uniform spraying; Step 703, system optimization and feedback loop: continuously monitor and evaluate the spraying effect and collect feedback data; use the feedback data to regularly update and optimize the machine learning model to improve the accuracy of the prediction and the precision of the spraying strategy.

[0026] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application uses a machine learning model to extract features from preprocessed leaf images, and combines environmental information data and crop growth conditions to build a training data set for model training, which achieves accurate prediction of crop health and water stress levels, and provides a reliable basis for formulating personalized spraying strategies. At the same time, through real-time data monitoring and dynamic adjustment, the system can adapt to various changes in the growth process of crops and improve the timeliness and accuracy of predictions. In addition, based on the prediction results and specific environmental information, the system can formulate accurate spraying strategies to meet the growth needs of crops and avoid problems such as excessive or insufficient spraying. Through continuous monitoring and evaluation of the spraying effect, and regular updating and optimization of the machine learning model using feedback data, the system can continuously adapt to new environments and needs, maintain its efficiency and stability, and realize the intelligence and performance optimization of the greenhouse automatic spraying control system.

[0027] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A greenhouse automatic sprinkler control system, characterized in that: include: Step 1, collecting indoor environmental information of the greenhouse through an information collection module, wherein the environmental information includes carbon dioxide concentration, air temperature and humidity, soil humidity, and light intensity; Step 2: Use a CMOS camera to collect crop images and capture their subtle features; Step 3, transmitting the crop image to the image processing system for image preprocessing, and obtaining the preprocessed crop image by grayscale processing, denoising and enhancing image features of the crop image; Step 4, performing status analysis on the preprocessed crop images to determine the health status of the crops or whether they are under water stress; Step 5: Use a camera to obtain an image of the crop height measuring scale, and obtain crop height data in real time through a combination of image processing and computer vision algorithms. The data is then transmitted to a control system in the small lifting device. The control system adjusts the position of the small lifting device according to the height data to ensure that the spray can cover all crops. Step 6: The intelligent decision-making and spraying control module comprehensively analyzes the crop health status, water stress level, environmental information and height data to formulate an appropriate spraying strategy; Step 7: The spraying equipment executes the spraying strategy to accurately spray the crops in the greenhouse to meet the moisture needs of the crops; Step 8: Build a training data set, train a learning model to predict the health status and water stress level of crops, deploy the learning model to the greenhouse automatic sprinkler control system, predict and formulate personalized sprinkler strategies in real time, continuously monitor the effects, collect feedback data, and optimize the model to improve accuracy.

2. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The information collection module includes a carbon dioxide concentration sensor, an air temperature and humidity sensor, a soil moisture sensor, and a light intensity sensor.

3. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The grayscale processing converts the color image into a grayscale image, thereby reducing the color information of the image.

4. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The denoising described above can smooth the image while maintaining the edge information of the image through Gaussian filtering and median filtering.

5. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The enhanced image features make the features of the crop images more prominent through edge detection operators, texture filters, sharpening filters and equalization of the grayscale histogram of the crop images.

6. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The health status of the crops is evaluated in the grayscale image by judging whether the grayscale values ​​of the crop leaves are concentrated between 100 and 150, whether the contrast is high, and whether the leaves are symmetrical left-right or top-bottom.

7. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The water stress can be determined by observing whether the edge of the leaf is no longer continuous or straight but has obvious bending or folding, whether the gray value distribution of the edge area has alternating light and dark stripes or block areas, and whether the leaf is wilted, the color becomes darker, and the glossiness decreases.

8. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The small lifting device is composed of a motor, a transmission system, a lifting platform and a control system. The motor provides power, the transmission system is responsible for converting the rotational motion of the motor into the vertical motion of the lifting platform, and the control system controls the operation of the motor according to the crop height data. The lifting platform can carry the spraying equipment up and down and adjust the position of the spraying equipment.

9. The automatic sprinkler control system for a greenhouse as claimed in claim 1, characterized in that: The training data set is formed by combining the health status, water stress level and environmental information of crops.

Images