Water conservancy irrigation intelligent monitoring method based on remote sensing data

Abstract

The invention discloses a water conservancy irrigation intelligent monitoring method based on remote sensing data, and relates to the technical field of water conservancy irrigation intelligent monitoring, and the method comprises the following steps: obtaining a soil moisture threshold value based on a first number of infrared images after irrigation under the same conditions with a to-be-monitored water conservancy irrigation area; obtaining an infrared abnormal region based on the infrared grey-scale map and a soil moisture threshold value; acquiring a light and shadow grey-scale map based on the regional light and shadow map and the infrared abnormal region; acquiring a first soil threshold value and a second soil threshold value based on a second number of normal images which are not irrigated under the same condition as the to-be-monitored water conservancy irrigation area; based on the shadow grey-scale map, the first soil threshold value and the second soil threshold value, obtaining an area needing to be irrigated; the method is used for solving the problem that an existing water conservancy irrigation intelligent monitoring technology cannot eliminate the influence of remote sensing data generated by interference objects in an irrigation area, and consequently the accuracy of a water conservancy irrigation intelligent monitoring result is low.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for water conservancy and irrigation, specifically to an intelligent monitoring method for water conservancy and irrigation based on remote sensing data. Background Technology

[0002] With the continuous development of agricultural and automation technologies, intelligent irrigation has become an important direction for modern agriculture. Irrigation monitoring primarily focuses on monitoring soil moisture content in the irrigated area to reflect irrigation effectiveness, allowing for timely control and adjustment. Traditional irrigation management relies mainly on manual field inspections, ground sensor deployment, and farmer self-reporting. While these methods can obtain soil moisture and irrigation information to some extent, they have significant limitations: they struggle to achieve large-scale, continuous, and comprehensive monitoring, failing to fully reflect spatial differences within the irrigation area; furthermore, the construction and maintenance of ground sensor networks are costly, and deployment in complex terrain is difficult; ultimately, they cannot meet the demands for high efficiency and ease of use in intelligent irrigation monitoring. With the rapid development of Earth observation technology, remote sensing technology has gained unique advantages such as macroscopic, objective, real-time, efficient, and low-cost operation. However, remote sensing technology cannot directly monitor humidity to determine whether irrigation requirements are met. Instead, it relies on indirect correlation analysis of remote sensing data to determine whether irrigation is needed. This indirect correlation introduces many interfering factors, leading to errors in irrigation monitoring data. For example, when monitoring irrigation based on infrared remote sensing images, the presence of piled-up hay, farm tools, or plastic film in the field—meaning interfering objects in the irrigation area—can easily lead to misjudgments of areas requiring irrigation. For instance, patent application CN117315457A discloses a smart irrigation monitoring method based on multi-source, multi-temporal remote sensing data. This method fails to eliminate remote sensing data generated by interfering objects in the irrigation area, resulting in errors in the monitoring results and hindering water conservation and crop growth. In other words, existing smart irrigation monitoring technologies fail to eliminate the influence of remote sensing data generated by interfering objects in the irrigation area, resulting in low accuracy of the smart irrigation monitoring results. Summary of the Invention

[0003] This invention aims to at least partially solve one of the technical problems in the prior art. It obtains a soil moisture threshold based on a first number of infrared images taken after irrigation under the same conditions as the irrigation area to be monitored; identifies infrared anomaly regions based on infrared grayscale images and soil moisture thresholds; obtains a light and shadow grayscale image based on regional light and shadow images and infrared anomaly regions; obtains a first soil threshold and a second soil threshold based on a second number of normal images taken under the same conditions as the irrigation area to be monitored but without irrigation; and identifies the area requiring irrigation based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold. This addresses the problem that existing intelligent irrigation monitoring technologies fail to eliminate the influence of remote sensing data generated by interfering objects in the irrigation area, resulting in low accuracy of intelligent irrigation monitoring results.

[0004] To achieve the above objectives, this application provides a method for intelligent monitoring of water conservancy and irrigation based on remote sensing data, comprising the following steps: Infrared remote sensing images of the white-heat model of the irrigation area to be monitored, obtained by UAV, are marked as regional infrared images; at the same time, normal images of the irrigation area are obtained and marked as regional light and shadow images. The infrared image of the region is converted to grayscale to obtain an infrared grayscale image. Soil moisture thresholds are obtained from infrared images of a first number of irrigation areas under the same conditions as the water conservancy irrigation areas to be monitored after irrigation. Infrared anomaly regions were obtained based on infrared grayscale images and soil moisture thresholds. Light and shadow grayscale images are obtained based on regional light and shadow maps and infrared anomaly regions; A first soil threshold and a second soil threshold are obtained based on a second number of normal images of the water conservancy irrigation area under the same conditions as the water conservancy irrigation area to be monitored that are not irrigated. The area requiring irrigation is determined based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold.

[0005] Further, the process of converting the regional infrared image to grayscale to obtain an infrared grayscale image includes the following sub-steps: Determine if the region's infrared image is a grayscale image. If it is, mark the region's infrared image as an infrared grayscale image. If not, obtain the RGB value of each pixel in the region's infrared image and mark it as an infrared RGB value. Convert the infrared RGB values ​​to grayscale values ​​using the grayscale conversion formula and mark them as infrared grayscale values. Convert the infrared RGB values ​​of each pixel in the region's infrared image to infrared grayscale values ​​to obtain an infrared grayscale image.

[0006] Furthermore, obtaining the soil moisture threshold based on the infrared images of a first number of moist soils under the same weather conditions as the irrigation area to be monitored includes the following sub-steps: The first number of infrared images obtained after irrigation under the same conditions as the water conservancy irrigation area to be monitored are marked as historical humid infrared images; the historical humid infrared images are then processed into grayscale to obtain historical humid grayscale images. Obtain the grayscale value of each pixel in the historical wet grayscale image and mark it as the historical wet grayscale value; Obtain the range of historical wet grayscale values, labeled A1 to A. k ;Transfer A1 to A k Divide the area evenly into c1 ranges, labeled as humidity segmentation ranges; the length of each humidity segmentation range is (A... k -A1)÷c1; The starting and ending values ​​of each humidity segment are A1+[(A k -A1)÷c1]×d1 and A1+[(A k -A1)÷c1]×(d1+1); where d1 is a positive integer from 0 to c1-1.

[0007] Furthermore, obtaining the soil moisture threshold based on the infrared images of a first number of moist soils under the same weather conditions as the irrigation area to be monitored also includes the following sub-steps: Count the frequency of historical humidity grayscale values ​​within each humidity segmentation range and label it as the humidity segmentation frequency; The humidity frequency threshold is calculated as: E1 = f1 × (H1 ÷ c1); where E1 is the humidity frequency threshold; f1 is the percentage of the humidity threshold; and H1 is the sum of all humidity frequency segments. Humidity segment frequencies that are less than the humidity frequency threshold are marked as abnormal segment frequencies; Sort the humidity segmentation frequency from left to right according to the starting value of the humidity segmentation range from smallest to largest; Determine whether the rightmost humidity segment frequency is an abnormal segment frequency. If so, delete the abnormal segment frequency and continue to determine whether the rightmost humidity segment frequency is an abnormal segment frequency after deleting the abnormal segment frequency, until the rightmost humidity segment frequency is no longer an abnormal segment frequency. When the rightmost humidity segment frequency is no longer an abnormal segment frequency, obtain the end value of the humidity segment range corresponding to the rightmost humidity segment frequency and mark it as the soil moisture threshold.

[0008] Furthermore, obtaining infrared anomaly regions based on infrared grayscale images and soil moisture thresholds includes the following sub-steps: The infrared grayscale values ​​in the infrared grayscale image that are greater than the soil moisture threshold are set to 0, and the infrared grayscale values ​​in the infrared grayscale image that are less than or equal to the soil moisture threshold are set to 255 to obtain a dry binarized image. The regions with a grayscale value of 0 in the dried binary image are marked as infrared anomalous regions.

[0009] Furthermore, obtaining the grayscale image of the light and shadow based on the regional light and shadow map and the infrared anomaly region includes the following sub-steps: Obtain the corresponding infrared anomalous region in the regional light and shadow map and mark it as the light and shadow anomalous region; convert the RGB value of each pixel in the light and shadow anomalous region into a grayscale value using a grayscale conversion formula to obtain the light and shadow grayscale map.

[0010] Furthermore, obtaining the first soil threshold and the second soil threshold based on a second number of normal images taken under the same conditions as the irrigation area to be monitored (without irrigation) includes the following sub-steps: The second set of normal images that were not irrigated under the same conditions as the water conservancy irrigation area to be monitored were marked as historical soil images; the historical soil images were processed into grayscale to obtain historical soil grayscale images; and the grayscale values ​​of the pixels in the historical soil grayscale images were marked as historical soil grayscale values.

[0011] Furthermore, obtaining the first soil threshold and the second soil threshold based on a second number of normal images taken under the same conditions as the irrigation area to be monitored but without irrigation also includes the following sub-steps: Obtain the range of historical soil gray values, marked as J1 to J n ; Transfer J1 to J n The soil is evenly divided into c2 regions, marked as soil subdivision regions; the length of each soil subdivision region is (J). n -J1)÷c2; The starting and ending values ​​of each soil segmentation range are J1+[(J n -J1)÷c2]×d2 and J1+[(J n -J1)÷c2]×(d2+1); where d2 is a positive integer from 0 to c2-1.

[0012] Furthermore, obtaining the first soil threshold and the second soil threshold based on a second number of normal images taken under the same conditions as the irrigation area to be monitored but without irrigation also includes the following sub-steps: The frequency of historical soil gray values ​​within each soil segment is counted and marked as the soil segment frequency. The soil frequency threshold is calculated as: E2 = f2 × (H2 ÷ c2); where E2 is the soil frequency threshold; f2 is the proportion of soil threshold; and H2 is the sum of the frequencies of all soil segments. Soil segment frequencies below the soil frequency threshold are marked as abnormal soil frequencies; Soil segmentation frequencies are sorted from left to right according to the starting value of the soil segmentation range, from smallest to largest. Determine whether the leftmost soil humidity segment frequency is an abnormal soil frequency. If so, delete the abnormal soil frequency and continue to determine whether the leftmost soil segment frequency after deleting the abnormal soil frequency is an abnormal soil frequency, until the leftmost soil segment frequency is no longer an abnormal soil frequency. When the leftmost soil segment frequency is no longer an abnormal soil frequency, obtain the starting value of the soil segment range corresponding to the leftmost soil segment frequency and mark it as the first soil threshold. Determine whether the rightmost soil humidity segment frequency is an abnormal soil frequency. If so, delete the abnormal soil frequency and continue to determine whether the rightmost soil segment frequency after deleting the abnormal soil frequency is an abnormal soil frequency, until the rightmost soil segment frequency is no longer an abnormal soil frequency. When the rightmost soil segment frequency is no longer an abnormal soil frequency, obtain the end value of the soil segment range corresponding to the rightmost soil segment frequency and mark it as the second soil threshold.

[0013] Furthermore, obtaining the area requiring irrigation based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold includes the following sub-steps: Set the gray values ​​in the light and shadow grayscale image that are less than the first soil threshold or greater than the second soil threshold to 255, and set the remaining gray values ​​to 0 to obtain the light and shadow binarized image. Obtain the regions with a grayscale value of 0 from the binary image of light and shadow, and mark them as areas that need to be irrigated; If an area is identified as needing irrigation, it indicates that the area still needs watering, and a signal will be sent indicating that watering is still required.

[0014] The beneficial effects of this invention are as follows: This invention obtains soil moisture thresholds based on a first number of infrared images after irrigation under the same conditions as the irrigation area to be monitored; identifies infrared anomaly areas based on infrared grayscale images and soil moisture thresholds; obtains light and shadow grayscale images based on regional light and shadow images and infrared anomaly areas; obtains a first soil threshold and a second soil threshold based on a second number of normal images of the irrigation area under the same conditions as the irrigation area to be monitored that have not been irrigated; and identifies the area requiring irrigation based on the light and shadow grayscale images, the first soil threshold, and the second soil threshold. The advantage is that it can eliminate the influence of remote sensing data generated by interfering objects in the irrigation area, thereby improving the accuracy of the identified area requiring irrigation. This invention obtains a first soil threshold and a second soil threshold based on a second number of normal images of the same irrigation area under the same conditions as the area to be monitored that are not irrigated. The advantage is that, due to the different colors of the soil and the interfering objects, the range of the obtained soil grayscale values ​​is from the first soil threshold to the second soil threshold. Therefore, the infrared abnormal areas can be further screened based on the first soil threshold and the second soil threshold, thus eliminating the influence of remote sensing data generated by interfering objects in the irrigation area and improving the accuracy of the obtained irrigation area. Attached Figure Description

[0015] Figure 1 This is a flowchart of the steps of the method of the present invention; Figure 2 This is a schematic diagram illustrating the frequency deletion of humidity segmentation according to the present invention; Figure 3 This is a schematic diagram of the infrared anomaly region of the present invention; Figure 4 This is a schematic diagram of soil segmentation frequency deletion according to the present invention; Figure 5 This is a schematic diagram of the area requiring irrigation according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1, please refer to Figure 1 As shown, this application provides a method for intelligent monitoring of water conservancy irrigation based on remote sensing data, including the following steps: Step S1: Based on the infrared remote sensing image of the white-heat model of the irrigation area to be monitored, which is obtained by the UAV, it is marked as the regional infrared image; at the same time, a normal image of the irrigation area is obtained and marked as the regional light and shadow image; here the irrigation area includes bare land areas. If it is completely covered by vegetation, it is difficult to identify it through the regional infrared image. When the irrigation area is completed, the temperature will drop. Therefore, the regional infrared image can be used to distinguish whether it has been irrigated; the regional light and shadow image is used to distinguish whether there are interfering objects.

[0018] Step S2 involves converting the regional infrared image to grayscale to obtain an infrared grayscale image; Step S2 includes the following sub-steps: Step S201: Determine whether the regional infrared image is a grayscale image. If yes, mark the regional infrared image as an infrared grayscale image; otherwise, obtain the RGB value of each pixel in the regional infrared image and mark it as an infrared RGB value; convert the infrared RGB values ​​to grayscale values ​​using a grayscale conversion formula and mark them as infrared grayscale values; convert the infrared RGB values ​​of each pixel in the regional infrared image to infrared grayscale values ​​to obtain an infrared grayscale image; the regional infrared image is an image in white-thermal mode, that is, the higher the temperature, the higher the brightness, similar to a grayscale image, but it may not be a grayscale image in grayscale format, so it needs to be converted to a grayscale image in a specific format; the grayscale conversion formula is to perform a weighted sum of the R, G, and B values ​​in the infrared RGB values; In practical applications, the regional infrared image is in RGB format. For example, if the infrared RGB value of a pixel in the regional infrared image is (52, 52, 52), then the infrared grayscale value of that pixel is (52+52+52)÷3=52. The infrared RGB value of each pixel in the regional infrared image is converted into an infrared grayscale value to obtain an infrared grayscale image.

[0019] Step S3 involves obtaining the soil moisture threshold based on a first number of infrared images obtained after irrigation under the same conditions as the irrigation area to be monitored. Step S3 includes the following sub-steps: Step S301: Mark the first number of infrared images obtained after irrigation under the same conditions as the water conservancy irrigation area to be monitored as historical humidity infrared images; perform grayscale processing on the historical humidity infrared images to obtain historical humidity grayscale images; it is necessary to meet the same conditions, such as temperature range, climate conditions, environment, etc., for comparison under the same conditions to be meaningful; the first number of historical humidity infrared images is used to obtain the range of historical humidity grayscale values; therefore, the larger the first number is set, the more accurate the range of historical humidity grayscale values ​​will be, but at the same time the amount of calculation will increase, so the first number should be set moderately, for example, the first number is 20; Step S302: Obtain the grayscale value of each pixel in the historical wet grayscale image and mark it as the historical wet grayscale value; Step S303: Obtain the range of historical wet grayscale values, marked as A1 to A... k ;Transfer A1 to A k Divide the area evenly into c1 ranges, labeled as humidity segmentation ranges; the length of each humidity segmentation range is (A... k -A1)÷c1; The starting and ending values ​​of each humidity segment are A1+[(A k -A1)÷c1]×d1 and A1+[(A k -A1)÷c1]×(d1+1); where d1 is a positive integer from 0 to c1-1; the humidity segmentation range is to better observe the distribution of historical humidity gray values; therefore, c1 should be set appropriately, for example, c1 is 10; d1 is a positive integer from 0 to 9; In practical applications, the range of historical humidity grayscale values ​​is 52 to 82. The range of 52 to 82 is evenly divided into 10 humidity segments, namely 52 to 55, 55 to 58, ..., 79 to 82; 52 and 55 are the starting and ending values ​​of the humidity segment when d1=0.

[0020] Step S304: Count the frequency of historical humidity grayscale values ​​within each humidity segmentation range and mark them as humidity segmentation frequencies; Step S305, calculate the humidity frequency threshold as: E1 = f1 × (H1 ÷ c1); where E1 is the humidity frequency threshold; f1 is the percentage of the humidity threshold; H1 is the sum of all humidity segment frequencies; the humidity frequency threshold is set to obtain the humidity segment range with a smaller humidity segment frequency. Since H1 ÷ c1 is the average humidity segment frequency for each humidity segment range, f1 is small, for example, f1 is 0.01. In practical applications, the sum of all humidity frequency segments is 41.4 million. For the sake of data convenience, the number of ten thousand is omitted, leaving only 4140. The humidity frequency threshold is calculated as: E1 = f1 × (H1 ÷ c1) = 0.01 × (4140 ÷ 10) = 4.14.

[0021] Step S306: Mark the humidity segmentation frequency that is less than the humidity frequency threshold as an abnormal segmentation frequency; Step S307: Sort the humidity segmentation frequency from left to right according to the starting value of the humidity segmentation range from small to large; Step S308: Determine whether the rightmost humidity segmentation frequency is an abnormal segmentation frequency. If so, delete the abnormal segmentation frequency and continue to determine whether the rightmost humidity segmentation frequency is an abnormal segmentation frequency after deleting the abnormal segmentation frequency, until the rightmost humidity segmentation frequency is no longer an abnormal segmentation frequency. When the rightmost humidity segmentation frequency is no longer an abnormal segmentation frequency, obtain the end value of the humidity segmentation range corresponding to the rightmost humidity segmentation frequency and mark it as the soil moisture threshold. In practical applications, humidity segmentation frequencies less than 4.14 are marked as abnormal segmentation frequencies. Please refer to [link / reference]. Figure 2 As shown, when the humidity segmentation frequency on the far right is not an abnormal segmentation frequency, 79 of the humidity segmentation range corresponding to 182 on the far right is obtained, and the soil moisture threshold is 79. Obtaining the soil moisture threshold excludes excessively large historical moisture gray values, thereby obtaining a more accurate distribution range of historical moisture gray values.

[0022] Step S4: Obtain infrared anomaly regions based on the infrared grayscale image and soil moisture threshold; Step S4 includes the following sub-steps: Step S401: Set the infrared grayscale values ​​in the infrared grayscale image that are greater than the soil moisture threshold to 0, and set the infrared grayscale values ​​in the infrared grayscale image that are less than or equal to the soil moisture threshold to 255 to obtain a dry binarized image; because the temperature decreases after watering the irrigated area, the grayscale value distribution range of the irrigated area after watering is low. If the grayscale value is high, it can be identified as an unirrigated area or an interference area. Step S402: Mark the areas with grayscale values ​​of 0 in the dry binarized image as infrared anomalous areas; infrared anomalous areas include un-irrigated areas and interference areas; In practical applications, infrared grayscale values ​​greater than 79 in the infrared grayscale image are set to 0, and infrared grayscale values ​​less than or equal to 79 in the infrared grayscale image are set to 255 to obtain a dry binarized image; please refer to Figure 3 for the obtained infrared anomaly area.

[0023] Step S5: Obtain a grayscale image of light and shadow based on the regional light and shadow map and the infrared anomaly region; Step S5 includes the following sub-steps: Step S501: Obtain the region corresponding to the infrared anomalous region in the regional light and shadow map and mark it as the light and shadow anomalous region; convert the RGB value of each pixel in the light and shadow anomalous region into a grayscale value using the grayscale conversion formula to obtain the light and shadow grayscale map; the conversion method is the same as that of the infrared grayscale map.

[0024] Step S6: Based on a second number of normal images taken under the same conditions as the irrigation area to be monitored but without irrigation, obtain a first soil threshold and a second soil threshold; Step S6 includes the following sub-steps: Step S601: Mark the second number of normal images of the irrigation area under the same conditions as the area to be monitored that have not been irrigated as historical soil images; perform grayscale processing on the historical soil images to obtain historical soil grayscale images; mark the grayscale values ​​of the pixels in the historical soil grayscale images as historical soil grayscale values; obtain the grayscale values ​​of the normally irrigated areas based on the normal light and shadow images. If the grayscale values ​​are not within the range of historical soil grayscale values, they can be identified as other objects, i.e., interference areas; the second number is 20 in order to obtain a more accurate range of historical soil grayscale values.

[0025] Step S602: Obtain the range of historical soil gray values, marked as J1 to J... n ; Transfer J1 to J n The soil is evenly divided into c2 regions, marked as soil subdivision regions; the length of each soil subdivision region is (J). n -J1)÷c2; The starting and ending values ​​of each soil segmentation range are J1+[(J n -J1)÷c2]×d2 and J1+[(J n -J1)÷c2]×(d2+1;where d2 is a positive integer from 0 to c2-1; the soil segmentation range is such that c2 is set appropriately to better observe the distribution of historical soil gray values, for example, c2 is 10; d2 is a positive integer from 0 to 9; In practical applications, the range of historical soil gray values ​​is 30 to 80; the range of 30 to 80 is evenly divided into 10 soil segmentation ranges; namely 30 to 35, 35 to 40, ..., 75 to 80; 30 to 35 are the starting and ending values ​​of the soil segmentation range when d1=0.

[0026] Step S603: Count the frequency of historical soil gray values ​​within each soil segmentation range and mark them as soil segmentation frequency; Step S604, calculate the soil frequency threshold as: E2 = f2 × (H2 ÷ c2); where E2 is the soil frequency threshold; f2 is the proportion of the soil threshold; H2 is the sum of the frequencies of all soil segments; the soil frequency threshold is set in order to obtain the soil segment range with a smaller soil segment frequency. Since H2 ÷ c2 is the average soil segment frequency of each soil segment range, f2 is set to be small, for example, f2 is 0.01. In practical applications, the sum of all soil segment frequencies is 41.4 million. For the sake of data convenience, the number of thousands will be omitted, leaving only 41.4 million. The humidity frequency threshold is then calculated as: E2 = f2 × (H2 ÷ c2) = 0.01 × (4140 ÷ 10) = 4.14. Step S605: Soil segment frequencies that are less than the soil frequency threshold are marked as abnormal soil frequencies; Step S606: Sort the soil segmentation frequency from left to right according to the starting value of the soil segmentation range from small to large; Step S607: Determine whether the leftmost soil humidity segmentation frequency is an abnormal soil frequency. If so, delete the abnormal soil frequency and continue to determine whether the leftmost soil segmentation frequency after deleting the abnormal soil frequency is an abnormal soil frequency, until the leftmost soil segmentation frequency is no longer an abnormal soil frequency. When the leftmost soil segmentation frequency is no longer an abnormal soil frequency, obtain the starting value of the soil segmentation range corresponding to the leftmost soil segmentation frequency and mark it as the first soil threshold. Step S608: Determine whether the rightmost humidity segmentation frequency is an abnormal soil frequency. If so, delete the abnormal soil frequency and continue to determine whether the rightmost soil segmentation frequency after deleting the abnormal soil frequency is an abnormal soil frequency, until the rightmost soil segmentation frequency is no longer an abnormal soil frequency. When the rightmost soil segmentation frequency is no longer an abnormal soil frequency, obtain the end value of the soil segmentation range corresponding to the rightmost soil segmentation frequency and mark it as the second soil threshold. The first soil threshold to the second soil threshold is the distribution range of gray values ​​in the irrigation area. In practical applications, soil fraction frequencies less than 4.14 are marked as anomalous soil frequencies; please refer to [link / reference]. Figure 4As shown, when the leftmost soil segment frequency is not an abnormal soil frequency, the starting value of the soil segment range corresponding to the leftmost soil segment frequency 182 is 35, so the first soil threshold is 35; when the rightmost soil segment frequency is not an abnormal soil frequency, the ending value of the soil segment range corresponding to the rightmost soil segment frequency 181 is 75, so the second soil threshold is 75; by obtaining the first soil threshold and the second soil threshold, excessively large or small historical soil gray values ​​are excluded, thereby obtaining a more accurate distribution range of historical soil gray values.

[0027] Step S7: Obtain the area requiring irrigation based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold; Step S7 includes the following sub-steps: Step S701: Set the gray values ​​in the light and shadow grayscale image that are less than the first soil threshold or greater than the second soil threshold to 255, and set the remaining gray values ​​to 0 to obtain a light and shadow binarized image; the distribution range of gray values ​​in the irrigation area from the first soil threshold to the second soil threshold is identified as the remaining objects, such as plastic bags. Step S702: Obtain the area composed of grayscale values ​​of 0 in the light and shadow binarized image and mark it as the area that needs to be watered; Step S703: If an area requiring irrigation is identified, it indicates that the area still needs watering, and a watering signal is issued. The area requiring irrigation is an un-irrigated area affected by remote sensing data generated by interfering objects; therefore, irrigation is required. This improves the accuracy of intelligent monitoring results for water conservancy irrigation. In practical applications, grayscale values ​​less than 35 or greater than 75 in the light and shadow grayscale image are set to 0, and the remaining grayscale values ​​are set to 255 to obtain a binary image of the light and shadow; please refer to [link / reference]. Figure 5 As shown, the irrigation area has been drawn.

[0028] Example 2: This application also provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions. The processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, steps such as those in the intelligent monitoring method for irrigation based on remote sensing data are performed to achieve the following functions: acquiring an infrared remote sensing image of the white-thermal model of the irrigation area to be monitored using a UAV, and marking it as a regional infrared image; simultaneously acquiring a normal image of the irrigation area, and marking it as a regional light and shadow image; performing grayscale processing on the regional infrared image to obtain an infrared grayscale image; obtaining a soil moisture threshold based on a first number of infrared images after irrigation under the same conditions as the irrigation area to be monitored; identifying infrared abnormal areas based on the infrared grayscale image and the soil moisture threshold; obtaining a light and shadow grayscale image based on the regional light and shadow image and the infrared abnormal areas; obtaining a first soil threshold and a second soil threshold based on a second number of normal images of the irrigation area to be monitored without irrigation under the same conditions; and identifying the area requiring irrigation based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold.

[0029] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0030] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the intelligent monitoring method for water conservancy irrigation based on remote sensing data provided by the above methods. The method includes: acquiring an infrared remote sensing image of a white-thermal model of the water conservancy irrigation area to be monitored based on a UAV, and marking it as a regional infrared image; simultaneously acquiring a normal image of the water conservancy irrigation area and marking it as a regional light and shadow image; performing grayscale processing on the regional infrared image to obtain an infrared grayscale image; obtaining a soil moisture threshold based on a first number of infrared images after irrigation under the same conditions as the water conservancy irrigation area to be monitored; obtaining an infrared abnormal area based on the infrared grayscale image and the soil moisture threshold; obtaining a light and shadow grayscale image based on the regional light and shadow image and the infrared abnormal area; obtaining a first soil threshold and a second soil threshold based on a second number of normal images of the water conservancy irrigation area to be monitored without irrigation; and obtaining the area to be irrigated based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold.

[0031] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the above-mentioned intelligent monitoring method for water conservancy irrigation based on remote sensing data to achieve the following functions: acquiring an infrared remote sensing image of the white-thermal model of the water conservancy irrigation area to be monitored based on a UAV, and marking it as a regional infrared image; simultaneously acquiring a normal image of the water conservancy irrigation area and marking it as a regional light and shadow image; performing grayscale processing on the regional infrared image to obtain an infrared grayscale image; obtaining a soil moisture threshold based on a first number of infrared images after irrigation under the same conditions as the water conservancy irrigation area to be monitored; obtaining an infrared abnormal area based on the infrared grayscale image and the soil moisture threshold; obtaining a light and shadow grayscale image based on the regional light and shadow image and the infrared abnormal area; obtaining a first soil threshold and a second soil threshold based on a second number of normal images without irrigation under the same conditions as the water conservancy irrigation area to be monitored; and obtaining the area to be irrigated based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold.

[0032] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0033] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An intelligent monitoring method for water conservancy irrigation based on remote sensing data, characterized in that, Comprising the following steps: Based on the unmanned aerial vehicle to obtain the water conservancy irrigation area to be monitored white-hot model of infrared remote sensing map, marked as regional infrared map; while obtaining the normal image of the water conservancy irrigation area, marked as regional light shadow map; The regional infrared map is grayed to obtain an infrared gray map; Based on the first number of the same conditions as the water conservancy irrigation area to be monitored, the soil wetting threshold is obtained based on the infrared image after irrigation; Based on the infrared gray map and the soil wetting threshold, the infrared abnormal area is obtained; Based on the regional light shadow map and the infrared abnormal area, a light shadow gray map is obtained; Based on the second number of the same conditions as the water conservancy irrigation area to be monitored, the first soil threshold and the second soil threshold are obtained based on the normal image without irrigation; Based on the light shadow gray map, the first soil threshold and the second soil threshold, the irrigation area is obtained.

2. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 1, characterized in that, The regional infrared map is grayed to obtain an infrared gray map, comprising the following sub-steps: Determine whether the regional infrared map is a gray map, if so, mark the regional infrared map as an infrared gray map; if not, obtain the RGB value of each pixel point in the regional infrared map, marked as infrared RGB value; the infrared RGB value is converted into a gray value by using a gray conversion formula, marked as infrared gray value; the infrared RGB value of each pixel point in the regional infrared map is converted into an infrared gray value, and an infrared gray map is obtained.

3. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 2, characterized in that, Based on the first number of the same conditions as the water conservancy irrigation area to be monitored, the soil wetting threshold is obtained based on the infrared image after irrigation; The first number of the same conditions as the water conservancy irrigation area to be monitored is marked as historical wetting infrared image; the historical wetting infrared image is grayed to obtain a historical wetting gray map; Obtain the gray value of each pixel point in the historical wetting gray map, marked as historical wetting gray value; A range of historical wetness gray scale values is obtained, marked as A1 to A k ; A1 to A k are evenly divided into c1 ranges, marked as humidity segmentation ranges; the range length of each humidity segmentation range is (A k -A1) ÷ c1; the starting value and the ending value of each humidity segmentation range are A1+[(A k -A1) ÷ c1]×d1 and A1+[(A k -A1) ÷ c1]×(d1+1) respectively; wherein d1 is a positive integer from 0 to c1-1.

4. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 3, characterized in that, Based on the first number of the same conditions as the water conservancy irrigation area to be monitored, the soil wetting threshold is obtained based on the infrared image after irrigation; The frequency of each humidity segmentation range is counted, marked as humidity segmentation frequency; The humidity frequency threshold is calculated as: E1=f1×(H1÷c1); wherein E1 is the humidity frequency threshold; f1 is the humidity threshold proportion; H1 is the sum of all humidity segmentation frequencies; The humidity segmentation frequency less than the humidity frequency threshold is marked as an abnormal segmentation frequency; The humidity segmentation frequency is sorted from left to right according to the starting value of the humidity segmentation range from small to large; Determine whether the rightmost humidity segmentation frequency is an abnormal segmentation frequency, if so, delete the abnormal segmentation frequency, continue to determine whether the rightmost humidity segmentation frequency after deleting the abnormal segmentation frequency is an abnormal segmentation frequency, until the rightmost humidity segmentation frequency is not an abnormal segmentation frequency; when the rightmost humidity segmentation frequency is not an abnormal segmentation frequency, the end value of the humidity segmentation range corresponding to the rightmost humidity segmentation frequency is obtained, marked as the soil wetting threshold.

5. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 4, characterized in that, Based on the infrared gray map and the soil wetting threshold, the infrared abnormal area is obtained; comprising the following sub-steps: The infrared grayscale values ​​in the infrared grayscale image that are greater than the soil moisture threshold are set to 0, and the infrared grayscale values ​​in the infrared grayscale image that are less than or equal to the soil moisture threshold are set to 255 to obtain a dry binarized image. The regions with a grayscale value of 0 in the dried binary image are marked as infrared anomalous regions.

6. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 5, characterized in that, Obtaining a grayscale image of light and shadow based on the regional light and shadow map and infrared anomaly region includes the following sub-steps: Obtain the corresponding infrared anomalous region in the regional light and shadow map and mark it as the light and shadow anomalous region; convert the RGB value of each pixel in the light and shadow anomalous region into a grayscale value using a grayscale conversion formula to obtain the light and shadow grayscale map.

7. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 6, characterized in that, Obtaining the first soil threshold and the second soil threshold based on a second number of normal images of the same unirrigated irrigation area as the area to be monitored includes the following sub-steps: The second set of normal images that were not irrigated under the same conditions as the water conservancy irrigation area to be monitored were marked as historical soil images; the historical soil images were processed into grayscale to obtain historical soil grayscale images; and the grayscale values ​​of the pixels in the historical soil grayscale images were marked as historical soil grayscale values.

8. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 7, characterized in that, Obtaining the first and second soil thresholds based on a second number of normal images of the same unirrigated irrigation area as the area to be monitored further includes the following sub-steps: obtain a range of historical soil gray values, marked as J1 to J n ; evenly divide J1 to J n into c2 ranges, marked as soil segmentation ranges; the range length of each soil segmentation range is (J n -J1) ÷ c2; the start value and end value of each soil segmentation range are J1+[(J n -J1) ÷ c2]×d2 and J1+[(J n -J1) ÷ c2]×(d2+1) respectively; wherein d2 is a positive integer from 0 to c2-1. 9.The method of claim 8, wherein, Obtaining the first and second soil thresholds based on a second number of normal images of the same unirrigated irrigation area as the area to be monitored further includes the following sub-steps: The frequency of historical soil gray values ​​within each soil segment is counted and marked as the soil segment frequency. The soil frequency threshold is calculated as: E2 = f2 × (H2 ÷ c2); where E2 is the soil frequency threshold; f2 is the proportion of soil threshold; and H2 is the sum of the frequencies of all soil segments. Soil segment frequencies below the soil frequency threshold are marked as abnormal soil frequencies; Soil segmentation frequencies are sorted from left to right according to the starting value of the soil segmentation range, from smallest to largest. Determine whether the leftmost soil humidity segment frequency is an abnormal soil frequency. If so, delete the abnormal soil frequency and continue to determine whether the leftmost soil segment frequency after deleting the abnormal soil frequency is an abnormal soil frequency, until the leftmost soil segment frequency is no longer an abnormal soil frequency. When the leftmost soil segment frequency is no longer an abnormal soil frequency, obtain the starting value of the soil segment range corresponding to the leftmost soil segment frequency and mark it as the first soil threshold. Determine whether the rightmost soil humidity segment frequency is an abnormal soil frequency. If so, delete the abnormal soil frequency and continue to determine whether the rightmost soil segment frequency after deleting the abnormal soil frequency is an abnormal soil frequency, until the rightmost soil segment frequency is no longer an abnormal soil frequency. When the rightmost soil segment frequency is no longer an abnormal soil frequency, obtain the end value of the soil segment range corresponding to the rightmost soil segment frequency and mark it as the second soil threshold.

10. The method for intelligent monitoring of water conservancy irrigation based on remote sensing data according to claim 9, characterized in that, The process of determining the area requiring irrigation based on the light and shadow grayscale image, the first soil threshold, and the second soil threshold includes the following sub-steps: Set the gray values ​​in the light and shadow grayscale image that are less than the first soil threshold or greater than the second soil threshold to 255, and set the remaining gray values ​​to 0 to obtain the light and shadow binarized image. Obtain the regions with a grayscale value of 0 from the binary image of light and shadow, and mark them as areas that need to be irrigated; If an area is identified as needing irrigation, it indicates that the area still needs watering, and a signal will be sent indicating that watering is still required.

Images