Sweet potato disease early warning device based on remote sensing technology

Abstract

The invention discloses a sweet potato disease early warning device based on a remote sensing technology, and relates to the technical field of crop planting, and the device comprises a processing module which divides a planting area into a plurality of recognition areas, and comprises a plurality of sub-processing units and an analysis unit; each sub-processing unit is matched with each identification area; the analysis unit is internally provided with a disease database, and the disease database is marked with a plurality of identification features based on the features of each sweet potato disease; the data monitoring module comprises a plurality of soil monitoring units; each soil monitoring unit is respectively matched with each identification area, monitors soil nutrient substances and soil environment in the corresponding identification area to obtain corresponding monitoring data, and transmits the monitoring data to the corresponding sub-processing unit; according to the method, the real-time remote sensing data of different identification areas are periodically detected, and disease symptoms can be timely found in the early stage of sweet potato diseases, so that early warning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of crop planting, and in particular to a sweet potato disease early warning device based on remote sensing technology. Background Art

[0002] Sweet potatoes are an important cash crop, widely cultivated worldwide and particularly valuable in China. Sweet potato yield and quality are affected by a variety of factors, with diseases being a major contributor to poor growth and yield reduction. Common sweet potato diseases include root rot, viral diseases, and anthracnose. These diseases typically manifest as visible lesions on the leaves, roots, or stems of affected plants, severely impacting sweet potato growth, quality, and yield.

[0003] With the development of remote sensing and the Internet of Things (IoT) technologies, real-time crop monitoring using drones and ground sensors has become a new trend in agricultural disease detection. Remote sensing technology captures high-precision image data of crop growth environments, enabling analysis of crop growth, nutritional requirements, and potential diseases.

[0004] After searching, a Chinese patent (publication number: CN107314816B) discloses a multi-level information monitoring and early warning method for early-stage crop diseases. The patent is to shoot the entire crop to obtain an infrared image of the monitored object; calculate the average temperature of the monitored object pixel points, the average temperature of the low-temperature pixel points, and the average temperature of the high-temperature pixel points respectively, and calculate the temperature difference between the average temperature of the high-temperature pixel points and the average temperature of the low-temperature pixel points; compare the temperature difference with the monitoring threshold and use it as the basis for determining whether the disease is infected; use a visible light-near-infrared hyperspectral imager to monitor diseased crops and obtain a characteristic image; use a microscopic imaging device to monitor the diseased spots in the characteristic image to obtain a microscopic image; and evaluate the extent of early-stage crop diseases based on the grayscale histogram of the microscopic image combined with the characteristic image.

[0005] Currently, traditional sweet potato disease detection methods rely primarily on manual testing and visual inspection. These methods require significant manpower and resources, and are difficult to implement in a timely and accurate manner across large-scale sweet potato cultivation areas. Furthermore, early disease identification is often limited by the severity of symptoms, making it difficult to implement effective prevention and control measures before the disease spreads, thereby impacting yield and quality. Therefore, this paper proposes a sweet potato disease early warning device based on remote sensing technology. Summary of the Invention

[0006] The purpose of the present invention is to provide a sweet potato disease early warning device based on remote sensing technology to solve the problems mentioned in the above background technology.

[0007] The present invention can be implemented through the following technical solutions: a sweet potato disease early warning device based on remote sensing technology, comprising a processing module, a data acquisition module, and a data monitoring module;

[0008] The processing module includes a plurality of sub-processing units and an analysis unit, and the processing module divides the planting area into a plurality of identification areas, and each sub-processing unit is matched with each identification area respectively;

[0009] The analysis unit has a built-in disease database, which is marked with multiple identification features based on the characteristics of various sweet potato diseases;

[0010] The data monitoring module includes a plurality of soil monitoring units, each of which is matched with each identification area and monitors the soil nutrients and soil environment in the corresponding identification area to obtain corresponding monitoring data, and transmits the monitoring data to the corresponding sub-processing unit;

[0011] Soil nutrients include nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, and other trace elements;

[0012] Soil environment includes soil pH, soil moisture, and soil temperature;

[0013] The data acquisition module periodically detects the planting area through remote sensing equipment to obtain remote sensing data, and the data acquisition module matches the remote sensing data with each identification area and transmits it to the corresponding sub-processing unit;

[0014] Remote sensing equipment uses drones, which are equipped with multispectral sensors and thermal infrared sensors;

[0015] Multispectral sensors are used to capture spectral information in multiple bands, thereby obtaining sweet potato vegetation index data for analyzing crop growth and health. When sweet potato health is poor, the vegetation index tends to drop significantly.

[0016] Thermal infrared sensors are used to obtain surface temperature data, analyze crop heat stress, and help identify water shortages, heat stress, or potential disease issues. When sweet potatoes are affected by diseases and pests, they often cause abnormal transpiration, which in turn affects surface temperature changes. Abnormal temperature changes can be detected by monitoring thermal data.

[0017] When using drones to inspect the planting area, multiple parallel flight paths are set up above the planting area, and the drones are flown back and forth on the flight paths to ensure that the entire planting area is covered to avoid data omissions;

[0018] Furthermore, when drones fly on adjacent flight paths, a 20% to 60% overlap is set to ensure that continuous images can be stitched together into a complete map.

[0019] Each sub-processing unit compares the monitoring data and remote sensing data it receives with the disease database in the analysis unit;

[0020] The analysis unit generates corresponding disease warning information based on the matched identification features.

[0021] A further technical improvement of the present invention is that: the analysis module establishes a monitoring map based on the geographical location of each identified area;

[0022] Each node in the monitoring graph corresponds to a recognition area, and each node is connected to the remaining recognition areas around the corresponding recognition area, and the edges between the nodes represent the difference data between adjacent recognition areas;

[0023] When the analysis module identifies the monitoring map, it detects the difference between each node and its adjacent identification area, and the analysis module presets a difference threshold θ. When the difference between a node and its adjacent identification area exceeds the difference threshold θ, the analysis module marks the identification area;

[0024] The formula used is: D ij =|X i -X j |;

[0025] Among them, D ij Represents the difference data between recognition area i and recognition area j;

[0026] X i 、X j are the monitoring data of identification area i and identification area j respectively;

[0027] and Warning i Indicates the marking device of the identification area i, 1 means marking, and 0 means not marking.

[0028] A further technical improvement of the present invention is that: the analysis module uses dynamic edges to monitor the monitoring graph;

[0029] Dynamic edges include user-defined edge lengths and random edge lengths. Different edge lengths represent the corresponding distances between the recognition area and its surrounding recognition areas.

[0030] Therefore, the difference data D between the recognition area i and the recognition area j is ij The formula used is:

[0031]

[0032] Among them, X i 、X j are the monitoring data of identification area i and identification area j respectively;

[0033] d ij is the interval between identification area i and identification area j, that is, the side length in the monitoring image;

[0034] α is an adjustment factor that controls the relationship between the weight and interval between recognition area i and recognition area j, and α>0.

[0035] A further technical improvement of the present invention is that: the analysis module analyzes the identifiable parts of each identification area based on remote sensing data, judges the proportion of vegetation index anomalies, temperature anomalies and the depth of anomalies, and assigns values ​​to the corresponding identification areas based on the judgment results;

[0036] The formula for the vegetation index anomaly depth is:

[0037]

[0038] in, is the vegetation index anomaly depth;

[0039] is the NDVI value of the jth pixel in the identified area i;

[0040] is the normal vegetation index value of the identified area i;

[0041] N i is the total number of vegetation index pixels in the identified area i;

[0042] The formula for the abnormal depth of temperature anomaly is:

[0043]

[0044] in, is the depth of temperature anomaly;

[0045] is the temperature value of the jth pixel in the identification area i;

[0046] is the normal temperature value of the identified area i;

[0047] Q i is the total number of temperature pixels in the identification area i;

[0048] The judgment formula is:

[0049]

[0050] V i It is the value assigned to the identification area, indicating the size of the healthy wind direction;

[0051] ω NDVI 、ω T are the weights of vegetation index anomaly and temperature anomaly, respectively;

[0052] are the proportions of vegetation index anomaly and temperature anomaly in the identifiable parts of the corresponding identification area i;

[0053] They are the depth of vegetation index anomaly and the depth of temperature anomaly;

[0054] The higher the proportion, the wider the abnormality range and the greater the risk;

[0055] The greater the depth, the more serious the anomaly and the greater the risk.

[0056] A further technical improvement of the present invention is that the analysis module allocates corresponding computing resources based on the assignment and marking status of the corresponding identification area to improve the comparison efficiency of the monitoring data and remote sensing data in the corresponding identification area with the disease database.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] By periodically monitoring real-time remote sensing data from different identified areas, the present invention can promptly detect signs of sweet potato diseases in their early stages, thereby achieving early warning. Furthermore, the present invention can accurately determine the likelihood of disease occurrence based on a comprehensive analysis of multi-dimensional data such as vegetation index, temperature anomalies, and soil nutrients, and issue early warnings to agricultural managers to prevent the spread of diseases and large-scale losses.

[0059] Furthermore, the data processing module intelligently analyzes remote sensing data with soil monitoring data. By establishing a disease database and combining disease characteristics and identification criteria, the system can automatically compare and accurately determine the disease type. Through pre-set thresholds and value assignment mechanisms, the system can generate timely and accurate disease warning information, providing a basis for decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0061] Figure 1 This is a system block diagram of the present invention. DETAILED DESCRIPTION

[0062] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0063] Example 1

[0064] See also Figure 1 As shown, the present invention provides a sweet potato disease early warning device based on remote sensing technology, including a processing module, a data acquisition module, and a data monitoring module;

[0065] The processing module includes a plurality of sub-processing units and an analysis unit, and the processing module divides the planting area into a plurality of identification areas, and each sub-processing unit is matched with each identification area respectively;

[0066] The analysis unit has a built-in disease database, which is marked with multiple identification features based on the characteristics of various sweet potato diseases;

[0067] The data monitoring module includes a plurality of soil monitoring units, each of which is matched with each identification area and monitors the soil nutrients and soil environment in the corresponding identification area to obtain corresponding monitoring data, and transmits the monitoring data to the corresponding sub-processing unit;

[0068] Soil nutrients include nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, and other trace elements;

[0069] Soil environment includes soil pH, soil moisture, and soil temperature;

[0070] The data acquisition module periodically detects the planting area through remote sensing equipment to obtain remote sensing data, and the data acquisition module matches the remote sensing data with each identification area and transmits it to the corresponding sub-processing unit;

[0071] In this embodiment, the remote sensing device adopts a drone, and the drone is equipped with a multispectral sensor and a thermal infrared sensor;

[0072] Multispectral sensors are used to capture spectral information in multiple bands, thereby obtaining sweet potato vegetation index data for analyzing crop growth and health. When sweet potato health is poor, the vegetation index tends to drop significantly.

[0073] Thermal infrared sensors are used to obtain surface temperature data, analyze crop heat stress, and help identify water shortages, heat stress, or potential disease issues. When sweet potatoes are affected by diseases and pests, they often cause abnormal transpiration, which in turn affects surface temperature changes. Abnormal temperature changes can be detected by monitoring thermal data.

[0074] When using drones to inspect the planting area, multiple parallel flight paths are set up above the planting area, and the drones are flown back and forth on the flight paths to ensure that the entire planting area is covered to avoid data omissions;

[0075] Furthermore, when the drones fly on adjacent flight paths, a 35% overlap of aerial photography is used in this embodiment to ensure that continuous images can be stitched together into a complete map;

[0076] The analysis module establishes a monitoring map based on the geographical location of each identified area;

[0077] Each node in the monitoring graph corresponds to a recognition area, and each node is connected to the remaining recognition areas around the corresponding recognition area, and the edges between the nodes represent the difference data between adjacent recognition areas;

[0078] When the analysis module identifies the monitoring map, it detects the difference between each node and its adjacent identification area, and the analysis module presets a difference threshold θ. When the difference between a node and its adjacent identification area exceeds the difference threshold θ, the analysis module marks the identification area;

[0079] The formula used is: D ij =|X i -X j |;

[0080] Among them, D ij Represents the difference data between recognition area i and recognition area j;

[0081] X i 、X j are the monitoring data of identification area i and identification area j respectively;

[0082] and Warning i Indicates the marking device of the identification area i, 1 means marking, and 0 means not marking.

[0083] Each sub-processing unit compares the monitoring data and remote sensing data it receives with the disease database in the analysis unit;

[0084] The analysis module analyzes the identifiable parts of each identification area based on remote sensing data, determines the proportion of vegetation index anomalies, temperature anomalies and the depth of anomalies, and assigns values ​​to the corresponding identification areas based on the judgment results;

[0085] The formula for the vegetation index anomaly depth is:

[0086]

[0087] in, is the vegetation index anomaly depth;

[0088] is the NDVI value of the jth pixel in the identified area i;

[0089] is the normal vegetation index value of the identified area i;

[0090] N i is the total number of vegetation index pixels in the identified area i;

[0091] The formula for the abnormal depth of temperature anomaly is:

[0092]

[0093] in, is the depth of temperature anomaly;

[0094] is the temperature value of the jth pixel in the identification area i;

[0095] is the normal temperature value of the identified area i;

[0096] Q i is the total number of temperature pixels in the identification area i;

[0097] The judgment formula is:

[0098]

[0099] V i It is the value assigned to the identification area, indicating the size of the healthy wind direction;

[0100] ω NDVI 、ω T are the weights of vegetation index anomaly and temperature anomaly, respectively;

[0101] are the proportions of vegetation index anomaly and temperature anomaly in the identifiable parts of the corresponding identification area i;

[0102] They are the depth of vegetation index anomaly and the depth of temperature anomaly;

[0103] The higher the proportion, the wider the abnormality range and the greater the risk;

[0104] The greater the depth, the more serious the anomaly and the greater the risk;

[0105] The analysis module allocates corresponding computing resources based on the assigned values ​​and marking status of the corresponding identified area to improve the efficiency of comparing the monitoring data and remote sensing data with the disease database in the corresponding identified area;

[0106] The analysis unit generates corresponding disease warning information based on the matched identification features.

[0107] Example 2

[0108] A sweet potato disease early warning device based on remote sensing technology, comprising a processing module, a data acquisition module, and a data monitoring module;

[0109] The processing module includes a plurality of sub-processing units and an analysis unit, and the processing module divides the planting area into a plurality of identification areas, and each sub-processing unit is matched with each identification area respectively;

[0110] The analysis unit has a built-in disease database, which is marked with multiple identification features based on the characteristics of various sweet potato diseases;

[0111] The data monitoring module includes a plurality of soil monitoring units, each of which is matched with each identification area and monitors the soil nutrients and soil environment in the corresponding identification area to obtain corresponding monitoring data, and transmits the monitoring data to the corresponding sub-processing unit;

[0112] Soil nutrients include nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, and other trace elements;

[0113] Soil environment includes soil pH, soil moisture, and soil temperature;

[0114] The data acquisition module periodically detects the planting area through remote sensing equipment to obtain remote sensing data, and the data acquisition module matches the remote sensing data with each identification area and transmits it to the corresponding sub-processing unit;

[0115] In this embodiment, the remote sensing device adopts a drone, and the drone is equipped with a multispectral sensor and a thermal infrared sensor;

[0116] Multispectral sensors are used to capture spectral information in multiple bands, thereby obtaining sweet potato vegetation index data for analyzing crop growth and health. When sweet potato health is poor, the vegetation index tends to drop significantly.

[0117] Thermal infrared sensors are used to obtain surface temperature data, analyze crop heat stress, and help identify water shortages, heat stress, or potential disease issues. When sweet potatoes are affected by diseases and pests, they often cause abnormal transpiration, which in turn affects surface temperature changes. Abnormal temperature changes can be detected by monitoring thermal data.

[0118] When using drones to inspect the planting area, multiple parallel flight paths are set up above the planting area, and the drones are flown back and forth on the flight paths to ensure that the entire planting area is covered to avoid data omissions;

[0119] Furthermore, when the drones fly on adjacent flight paths, a 40% overlap of aerial photography is used in this embodiment to ensure that continuous images can be stitched together into a complete map;

[0120] Each sub-processing unit compares the monitoring data and remote sensing data it receives with the disease database in the analysis unit;

[0121] The analysis module establishes a monitoring map based on the geographical location of each identified area;

[0122] Compared with Example 1, the analysis module in Example 2 uses dynamic edges to monitor the monitoring graph;

[0123] Dynamic edges include user-defined edge lengths and random edge lengths. Different edge lengths represent the corresponding distances between the recognition area and its surrounding recognition areas.

[0124] Therefore, the difference data D between the recognition area i and the recognition area j is ij The formula used is:

[0125]

[0126] Among them, X i 、X j are the monitoring data of identification area i and identification area j respectively;

[0127] d ij is the interval between identification area i and identification area j, that is, the side length in the monitoring image;

[0128] α is an adjustment factor that controls the relationship between the weight and interval between recognition area i and recognition area j, and α>0.

[0129] The analysis module analyzes the identifiable parts of each identification area based on remote sensing data, determines the proportion of vegetation index anomalies, temperature anomalies and the depth of anomalies, and assigns values ​​to the corresponding identification areas based on the judgment results;

[0130] The formula for the vegetation index anomaly depth is:

[0131]

[0132] The formula for the anomaly depth of temperature anomaly is:

[0133]

[0134] The judgment formula is:

[0135]

[0136] The analysis unit generates corresponding disease warning information based on the matched identification features.

[0137] Specific examples:

[0138] In both sets of embodiments, the disease data includes the following features:

[0139] 1. Root rot, which mainly affects the roots and manifests as slowed growth;

[0140] Vegetation index characteristics: low NDVI value, high negative change;

[0141] Temperature anomaly characteristics: abnormally high temperature;

[0142] Soil characteristics: low nitrogen and low phosphorus;

[0143] 2. Virus disease: spots appear on the leaf surface of diseased sweet potato plants;

[0144] Vegetation index characteristics: medium NDVI value, slight fluctuation;

[0145] Temperature anomaly characteristics: small temperature difference;

[0146] Soil characteristics: high sulfur, low potassium;

[0147] 3. Anthracnose, which causes leaf wilt and spreads rapidly;

[0148] Vegetation index characteristics: extremely low NDVI value, with a sharp decline;

[0149] Temperature anomaly characteristics: abnormally high temperature;

[0150] Soil characteristics: high potassium, low calcium.

[0151] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A sweet potato disease early warning device based on remote sensing technology, characterized in that: include: a processing module, which divides the planting area into a plurality of identification areas and includes a plurality of sub-processing units and an analysis unit; Each sub-processing unit is matched with each recognition area respectively; The analysis unit has a built-in disease database, which is marked with multiple identification features based on the characteristics of various sweet potato diseases; Data monitoring module, including multiple soil monitoring units; Each soil monitoring unit is matched with each identification area, and monitors the soil nutrients and soil environment in the corresponding identification area to obtain corresponding monitoring data, and transmits the monitoring data to the corresponding sub-processing unit; The data acquisition module periodically detects the planting area through remote sensing equipment to obtain remote sensing data, and the data acquisition module matches the remote sensing data with each identification area and transmits it to the corresponding sub-processing unit; Each sub-processing unit compares the monitoring data and remote sensing data it receives with the disease database in the analysis unit; The analysis unit generates corresponding disease warning information based on the matched identification features.

2. The sweet potato disease early warning device based on remote sensing technology according to claim 1, characterized in that: Remote sensing equipment uses drones, which are equipped with multispectral sensors and thermal infrared sensors; The multispectral sensor captures spectral information in multiple bands and obtains vegetation index data of sweet potatoes; Thermal infrared sensors are used to obtain surface temperature data.

3. The sweet potato disease early warning device based on remote sensing technology according to claim 2, characterized in that: Multiple parallel flight paths are set up on the upper side of the planting area; The drone flies back and forth on the flight path, covering the entire planting area; In addition, when the drones fly on adjacent flight paths, an aerial photography overlap of 20% to 60% is set.

4. The sweet potato disease early warning device based on remote sensing technology according to claim 3, characterized in that: The analysis module establishes a monitoring map based on the geographical location of each identified area; When the analysis module identifies the monitoring map, the difference between each node and its adjacent identification area is detected, and the analysis module is preset with a difference threshold; When the difference between a node and its adjacent identification area exceeds a difference threshold, the analysis module marks the identification area.

5. The sweet potato disease early warning device based on remote sensing technology according to claim 4, characterized in that: The analysis module monitors the monitoring graph using dynamic edges; Dynamic edges include user-defined edge lengths and random edge lengths. Different edge lengths represent the corresponding distances between the recognition area and its surrounding recognition areas.

6. A sweet potato disease early warning device based on remote sensing technology according to claim 4 or 5, characterized in that: The analysis module analyzes the identifiable parts of each identification area based on remote sensing data, judges the proportion and depth of vegetation index anomalies and temperature anomalies, and assigns values ​​to the corresponding identification areas based on the judgment results.

7. The sweet potato disease early warning device based on remote sensing technology according to claim 6, characterized in that: The formula for the vegetation index anomaly depth is: Among them, Z i NDVI is the vegetation index anomaly depth; is the NDVI value of the jth pixel in the identified area i; is the normal vegetation index value of the identified area i; N i is the total number of vegetation index pixels in the identified area i; The formula for the abnormal depth of temperature anomaly is: Among them, W i T is the depth of temperature anomaly; is the temperature value of the jth pixel in the identification area i; is the normal temperature value of the identified area i; Q i is the total number of temperature pixels in the identification area i; The judgment formula is: V i It is the value assigned to the identification area, indicating the size of the healthy wind direction; ω NDVI 、ω T are the weights of vegetation index anomaly and temperature anomaly, respectively; P i NDVI 、P i T are the proportions of vegetation index anomaly and temperature anomaly in the identifiable parts of the corresponding identification area i; W i T They are the vegetation index anomaly depth and temperature anomaly depth respectively.

8. The sweet potato disease early warning device based on remote sensing technology according to claim 7, characterized in that: The analysis module allocates corresponding computing resources based on the assigned values ​​and marking states of the corresponding identification areas.

Images