Forestry pest and disease damage intelligent monitoring and early warning system based on Internet of Things

Abstract

The invention discloses a forestry pest and disease damage intelligent monitoring and early warning system based on the Internet of Things, and relates to the technical field of pest and disease damage management. The forestry pest and disease damage intelligent monitoring and early warning system based on the Internet of Things comprises a data acquisition module used for collecting an environment characteristic data set in real time; the preprocessing module is used for carrying out data vectorization, data cleaning, data synchronization and data standardization processing on the environment characteristic data set collected in real time; the transmission module is used for selecting an optimal data transmission path for data transmission according to the node state analysis result; and the abnormity checking module is used for carrying out data abnormity detection on the real-time environment characteristic data set transmitted by the transmission module. The problem that in an existing forestry pest and disease damage intelligent monitoring and early warning system based on the Internet of Things in the field environment, equipment may be interfered by non-pest and disease damage factors, a sensor may be touched by mistake, and false alarm is caused is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of pest management, and in particular to an intelligent monitoring and early warning system for forest pests based on the Internet of Things. Background Art

[0002] In the field of forestry management and plant pest and disease monitoring, traditional monitoring methods rely on manual inspections and fixed-point observations, which are labor-intensive, inefficient, and slow to respond. With the development of the forestry economy and the impact of global climate change, the frequent occurrence of forest pests and diseases has become a major challenge in global forest resource management. This situation requires more efficient and automated solutions to deal with pest and disease monitoring and early warning.

[0003] With the development of Internet of Things technology, its application in environmental monitoring and ecological management has gradually increased, providing new technical support for the monitoring of forest pests and diseases. Using Internet of Things devices, such as various environmental sensors and image capture devices, detailed data about the forest environment can be collected in real time and sent to the central server through wireless networks for processing and analysis. However, the existing system still has shortcomings in data processing efficiency, real-time performance, and accuracy. Especially in the field environment, the equipment is easily disturbed by non-pest and disease factors such as wildlife activities and extreme weather, which increases the risk of false alarms.

[0004] Therefore, in response to the above problems, an intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things is urgently needed. Summary of the invention

[0005] Technical issues solved

[0006] In view of the deficiencies in the prior art, the present invention provides an intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things, which solves the problem that in the existing intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things in the wild environment, the equipment may be interfered by non-pest and disease factors, which may accidentally touch the sensor and cause false alarms.

[0007] Technical Solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things, including a data acquisition module for rationally planning and arranging the positions of sensors according to the terrain of the forest area, the type of vegetation and the historical occurrence of pests and diseases, and collecting environmental feature data sets in real time; a preprocessing module for performing data vectorization, data cleaning, data synchronization and data standardization processing on the environmental feature data sets collected in real time; a transmission module for real-time collection and analysis of the status of transmission nodes, and selecting the best path for data transmission according to the results of node status analysis for data transmission; an anomaly detection module for performing data anomaly detection on the real-time environmental feature data sets transmitted through the transmission module, and performing anomaly processing according to the results of data anomaly detection, storing the normal environmental feature data sets, marking them as historical environmental feature data sets, obtaining historical pest and disease signal data and historical interference signal data, and establishing an environmental feature database; an anti-interference prediction processing module for performing interference detection on the normal environmental feature data sets processed by the anomaly detection module, and performing signal filtering according to the interference detection results; a real-time early warning module for monitoring pest and disease activity according to real-time data conditions, and taking corresponding processing measures.

[0009] Furthermore, the specific process of real-time collection of environmental feature data sets is as follows: the environmental feature data sets include air temperature, air humidity, light intensity, vibration data, image data and sound signals; using a temperature sensor to measure air temperature, using a temperature sensor to measure air humidity, using a light sensor to measure light intensity, using a vibration sensor to detect tiny vibrations of tree trunks and branches and leaves and marking them as vibration data, using a high-definition camera to obtain images of plant surfaces and marking them as image data, and using a microphone to collect sounds such as insect sounds and marking them as sound signals.

[0010] Furthermore, the specific process of data vectorization, data cleaning, data synchronization and data standardization of the environmental feature data set collected in real time is as follows: data vectorization: converting all collected environmental feature data sets into numerical vector form; organizing air temperature, air humidity, light intensity, vibration data, image data and sound signals into multidimensional feature vectors, each vector represents a complete data record at a time point, respectively marked as temperature feature vector, humidity feature vector, light intensity feature vector, vibration feature vector, image feature vector and sound feature vector; data cleaning: detecting and processing missing values, outliers and noise data in the environmental feature data set; data synchronization: synchronizing data from different sensors according to timestamps; data standardization: normalizing the data in the environmental feature data set to eliminate dimensional differences.

[0011] Furthermore, the real-time collection and analysis of the status of the transmission nodes and the selection of the best path for data transmission according to the node status analysis results are as follows: node initialization: when each sensor node is started, it broadcasts its own status information, including remaining energy, location information and environmental parameters; neighbor discovery: the node establishes a neighbor list by receiving the broadcast information of the neighboring nodes to prepare for path selection; path selection: obtain the number of hops in the transmission path, obtain the received signal strength, the remaining energy of the node, the node temperature, the node humidity and the node light intensity; the best path selection value is comprehensively calculated through the number of hops in the path, the received signal strength, the remaining energy of the node, the node temperature, the node humidity and the node light intensity, and according to the calculation result, the path with the smallest best path selection value is selected as the data transmission path.

[0012] Furthermore, the specific calculation formula of the optimal path selection value is: In the formula, C path represents the optimal path selection value, n represents the number of hops in the path, HC i represents the number of hops in the i-th segment, w1 represents the weight coefficient of the number of hops in the i-th segment, RSSI i represents the received signal strength of the i-th segment, w2 represents the weight coefficient of the received signal strength of the i-th segment, E i represents the residual energy of the ith node, w3 represents the weight coefficient of the residual energy of the ith node, T i represents the node temperature of the i-th node, a represents the influence coefficient of the i-th node temperature, H i represents the node humidity of the i-th node, b represents the influence coefficient of the i-th node humidity, L i represents the node illumination intensity of the i-th node, c represents the influence coefficient of the i-th node illumination intensity, and w4 represents the weight coefficient of the environmental factor.

[0013] Furthermore, the real-time environmental feature data set transmitted through the transmission module is subjected to data anomaly detection, and the specific process of anomaly processing is as follows: obtaining a multidimensional feature vector of the current data point, obtaining a mean vector and a covariance matrix from a historical environmental feature data set in an environmental feature database, and obtaining a data anomaly detection value by comprehensively calculating the multidimensional feature vector, the mean vector and the covariance matrix; comparing the data anomaly detection value with a first threshold in real time, and when the data anomaly detection value is less than or equal to the first threshold, no processing is performed, and when the data anomaly detection value is greater than the first threshold, it is determined to be abnormal data, and the abnormal data is removed from the analysis.

[0014] Furthermore, the specific calculation formula of the data anomaly detection value is: In the formula, AD represents the data anomaly detection value, x represents the multidimensional feature vector of the current data point, μ represents the mean vector, and ∑ represents the covariance matrix.

[0015] Furthermore, the interference detection is performed on the normal environmental feature data set processed by the abnormality detection module, and the signal filtering is performed according to the interference detection result. The specific process is: obtaining sound feature vectors, historical pest and disease signal data and historical interference signal data, calculating the probability density value of historical pest and disease signal data and the probability density value of historical interference signal data, and obtaining the interference data detection value by comprehensive calculation of the sound feature vectors, the probability density value of historical pest and disease signal data and the probability density value of historical interference signal data; comparing the interference data detection value and the second threshold in real time, when the interference data detection value is less than or equal to the second threshold, it is judged to be an interference signal, and the data judged to be an interference signal is intercepted and does not enter the next link; when the interference data detection value is greater than the second threshold, it is judged to be pest and disease signal data, and the data judged to be a pest and disease signal is retained for subsequent analysis.

[0016] Furthermore, the specific calculation formula of the interference data detection value is: Where IDV represents the interference data detection value, P(S|Z0) represents the probability density value of historical pest and disease signal data, and P(S|Q1) represents the probability density value of historical interference signal data.

[0017] Furthermore, the specific process of monitoring pest and disease activity according to real-time data and taking corresponding treatment measures is: receiving and analyzing pest and disease signal data processed by the anti-interference prediction processing module in real time, detecting whether there is abnormal pest and disease activity according to the real-time pest and disease signal data, and if abnormal pest and disease activity is detected, triggering an alarm and notifying relevant personnel.

[0018] Beneficial Effects

[0019] The present invention has the following beneficial effects:

[0020] (1) The present invention uses a variety of sensors to collect environmental characteristic data sets in real time, including air temperature, humidity, light intensity, vibration, image and sound data, and performs advanced preprocessing and analysis on these data. The system can more accurately detect signs of pests and diseases. This further improves the monitoring accuracy and effectively solves the problem of misjudgment caused by single data or insufficient processing in the prior art.

[0021] (2) The present invention analyzes the status of the transmission node, and the system can select the best data transmission path, which not only optimizes the data transmission efficiency, but also ensures the stability and reliability of data transmission. It also realizes the rational use of network resources and effectively solves the problem of data transmission delay or loss caused by network problems in the prior art.

[0022] (3) In the present invention, through the abnormality detection module and the anti-interference prediction processing module, the system can identify and filter out interference signals caused by non-pest factors, such as natural noise and the influence of non-target biological activities. This further achieves the accuracy of monitoring data and the reliability of the system, and effectively solves the problem of false alarms caused by environmental noise and interference in the prior art.

[0023] (4) The present invention uses a real-time early warning module to monitor pest activities based on the data obtained through analysis. Once abnormal activities are detected, an alarm is immediately triggered and relevant personnel are notified. This real-time response mechanism greatly shortens the response time, thereby achieving effective protection of forestry resources and effectively solving the problem of increased potential losses caused by slow response in the prior art.

[0024] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a structural diagram of the forestry pest intelligent monitoring and early warning system based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] See also Figure 1The embodiment of the present invention provides a technical solution: an intelligent monitoring and early warning system for forest pests and diseases based on the Internet of Things, including a data acquisition module for rationally planning and arranging sensor locations according to forest terrain, vegetation types and historical pest and disease occurrences, and collecting environmental feature data sets in real time; a preprocessing module for performing data vectorization, data cleaning, data synchronization and data standardization processing on the environmental feature data sets collected in real time; a transmission module for real-time collection and analysis of the status of transmission nodes, and selecting the best path for data transmission according to the node status analysis results for data transmission; an anomaly detection module for performing data anomaly detection on the real-time environmental feature data sets transmitted through the transmission module, and performing anomaly processing according to the results of the data anomaly detection, storing the normal environmental feature data sets, marking them as historical environmental feature data sets, obtaining historical pest and disease signal data and historical interference signal data, and establishing an environmental feature database; an anti-interference prediction processing module for performing interference detection on the normal environmental feature data sets processed by the anomaly detection module, and performing signal filtering according to the interference detection results; a real-time early warning module for monitoring pest and disease activity according to real-time data conditions, and taking corresponding processing measures.

[0028] Specifically, the specific process of real-time collection of environmental feature data sets is as follows: environmental feature data sets include air temperature, air humidity, light intensity, vibration data, image data and sound signals; during the monitoring process, high-precision temperature sensors are used to measure air temperature, which helps analyze the potential impact of temperature changes on forest ecology. The same sensor is also used to measure air humidity, and humidity levels are particularly critical for assessing the risk of plant diseases. Light sensors are responsible for measuring light intensity, which is one of the important factors for plant growth and the development of pests and diseases. Vibration sensors are installed on tree trunks and branches to detect tiny vibrations caused by wind or biological activities such as insect activity and bird perching. These data are marked as vibration data. High-definition cameras are used to obtain images of plant surfaces. These image data are very valuable for identifying specific symptoms of pests and diseases. In addition, microphones are used to collect sounds in the environment, such as natural sounds such as insect sounds. These sound signals may indicate the presence of certain specific pests and diseases.

[0029] In this embodiment, the process of real-time collection of environmental characteristic data sets effectively enhances the comprehensive perception capability of the forestry pest and disease monitoring system by using a variety of sensor systems to comprehensively monitor various environmental parameters of the forest. Through temperature and humidity sensors, the system can obtain accurate information about climate conditions, which is essential for assessing the development potential of pests and diseases. The measurement of light intensity provides data on biological activity and the impact of photosynthesis, while the tiny vibrations and sound signals captured by vibration sensors and microphones help to detect activities caused by insects or other small animals early. In addition, the image data obtained by high-definition cameras makes it possible to intuitively identify pathological changes and external damage on plants. The integration of these data not only improves the accuracy of monitoring, but also greatly enhances the response speed and efficiency of the early warning system, making pest and disease management more scientific and systematic, thereby achieving effective protection of forest resources and a significant reduction in losses from pests and diseases.

[0030] Specifically, the real-time collected environmental feature data sets are processed by data vectorization, data cleaning, data synchronization and data standardization. The specific process is as follows: Data vectorization: All collected environmental feature data sets are converted into numerical vector form. This includes organizing information such as air temperature, air humidity, light intensity, vibration data, image data and sound signals into multi-dimensional feature vectors. Each vector represents a complete data record at a time point and is marked as a temperature feature vector, a humidity feature vector, a light intensity feature vector, a vibration feature vector, an image feature vector and a sound feature vector. This vectorization process not only facilitates the subsequent processing and analysis of the data, but also helps the machine learning model to interpret and utilize the data more effectively. Data cleaning: The data in the environmental feature data set is detected to identify missing values, outliers and noise data, and then processed accordingly. This step is key to ensuring data accuracy and reliability, because dirty or erroneous data will directly affect the accuracy of monitoring results and the effectiveness of decision-making. Data synchronization: Data collected by different sensors are synchronized according to timestamps. This ensures that the data at each time point in the data set comes from the same time frame, thereby eliminating analysis errors caused by data asynchrony. Data synchronization is particularly important for time series analysis and real-time monitoring, which helps the system accurately track environmental changes and the development of pests and diseases. Data standardization: Normalize the data in the environmental characteristic data set to eliminate dimensional differences. Data standardization is to reduce the impact of different dimensions and magnitudes, so that data from different sensors and different ranges can be compared and analyzed under the same standards.

[0031] In this implementation scheme, by performing data vectorization, data cleaning, data synchronization and data standardization on the environmental feature data set collected in real time, this system greatly improves the quality and consistency of the data, and provides a solid foundation for the accuracy and efficiency of the intelligent monitoring and early warning system for forest pests and diseases. Data vectorization ensures that various sensor data can be effectively integrated and analyzed, while data cleaning removes outliers and noise that may interfere with the analysis results, ensuring the purity and availability of the data. Data synchronization eliminates analysis errors that may be caused by time dislocation, and ensures that all data are compared and analyzed based on a unified timeline. Finally, data standardization eliminates potential deviations between different sensors and different data ranges, making the data more standardized and convenient for analysis across time and space. The comprehensive execution of these steps effectively supports the system to carry out efficient and accurate pest and disease monitoring and early warning, and further realizes the effective protection of forest resources and the reduction of pest and disease losses.

[0032] Specifically, the status of the transmission nodes is collected and analyzed in real time, and the best path for data transmission is selected according to the results of the node status analysis. The specific process is as follows: Node initialization: In the early stage of system deployment, each sensor node broadcasts its own status information when it starts, including the node's remaining energy, precise location information, and current environmental parameters. This step not only helps the system administrator monitor the operating status of each node, but also serves as the basis for subsequent data transmission routing selection. Neighbor discovery: Each node listens to the broadcast information from the surrounding neighbor nodes and establishes a neighbor list through this information. This list records the status of each adjacent node in detail, including its energy level and signal quality, providing necessary data support for the subsequent path selection. Path selection: In this stage, the system calculates potential paths for data transmission. First, the system obtains relevant data for each hop in the transmission path, including the number of hops, the received signal strength, the node's remaining energy, and the node's environmental parameters such as temperature, humidity, and light intensity. Then, the system uses this data to calculate the overall score of each path, which takes into account factors such as path stability, transmission efficiency, and energy consumption minimization. The path with the lowest score is considered to be the best data transmission path.

[0033] In this implementation scheme, real-time collection and analysis of the status of transmission nodes and selection of the best path for data transmission based on these analysis results are key measures to improve the efficiency and reliability of data transmission. Through node initialization, the system is able to monitor the operating status of each sensor node in real time, including its remaining energy and environmental conditions, thereby ensuring the optimal operation and data transmission of the node. The neighbor discovery process enables each node to understand the status of its surrounding nodes, providing the necessary information for intelligent path selection. Finally, through the comprehensive evaluation of various data such as signal strength, node energy, environmental parameters, etc., the system is able to select the best data transmission path. This approach not only significantly reduces delays and losses in data transmission, but also improves the system's adaptability to environmental changes and the stability of the overall network. In addition, optimized path selection reduces energy consumption, prolongs the operation time of nodes, and further enhances the durability and reliability of the system. This series of measures greatly improves the real-time response capability and accuracy of the forestry pest and disease monitoring system, and effectively supports the decision-making process of forestry protection and management.

[0034] Specifically, the specific calculation formula of the optimal path selection value is: In the formula, C path represents the optimal path selection value, n represents the number of hops in the path, HC i Indicates the number of hops in the i-th segment, which is directly obtained from the network topology and is a basic network parameter. w1 indicates the weight coefficient of the hop count in the i-th segment, which usually ranges from 0.1 to 0.5 and is set according to the complexity of the network topology and the importance of the hop count. RSSI i represents the received signal strength of the i-th segment, which is directly measured by the communication hardware of the node. w2 represents the weight coefficient of the received signal strength of the i-th segment, and its value range is usually 0.2 to 0.6, which is adjusted according to the role of signal strength in network stability. i represents the residual energy of the ith node, which is directly read from the node's energy management system. w3 represents the weight coefficient of the residual energy of the ith node, ranging from 0.1 to 0.4, reflecting the importance of energy efficiency in path selection. i represents the node temperature of the i-th node, a represents the influence coefficient of the i-th node temperature, H i represents the node humidity of the i-th node, b represents the influence coefficient of the i-th node humidity, L irepresents the node illumination intensity of the i-th node, c represents the influence coefficient of the illumination intensity of the i-th node, a, b, c are the influence coefficients of environmental factors (temperature, humidity, illumination), usually ranging from 0.05 to 0.3, which are set according to the variability of actual environmental conditions and the degree of influence on network performance, w4 represents the weight coefficient of environmental factors, ranging from 0.2 to 0.5, representing the comprehensive influence weight of environmental variables in path selection, and the weight coefficient and influence coefficient are obtained from the environmental feature database.

[0035] In this implementation scheme, by implementing the optimal path selection algorithm, the intelligent monitoring and early warning system for forest pests and diseases based on the Internet of Things can significantly improve the efficiency and reliability of data transmission. The algorithm comprehensively considers multiple key parameters, including the number of node hops, signal strength, residual energy, and environmental factors such as temperature, humidity, and light intensity, so as to select the transmission path with the lowest cost under various network conditions and environmental settings. This not only optimizes the use of network resources, reduces the delay and packet loss rate of data transmission, but also improves the response speed and accuracy of the entire monitoring system. In addition, through intelligent path selection, the system can adapt to environmental changes and network status fluctuations, ensuring that monitoring data can be reliably transmitted to the analysis center under various conditions, thereby effectively supporting real-time monitoring and timely early warning of forest pests and diseases. This step is the key to ensuring the overall performance and effectiveness of the system, and helps to improve the operational efficiency and long-term sustainability of the system.

[0036] Specifically, data anomaly detection is performed on the real-time environmental feature data set transmitted by the transmission module, and anomaly processing is performed according to the result of data anomaly detection. The specific process is as follows: first, a multidimensional feature vector of the current data point is obtained from the received data. These feature vectors may include data collected from various environmental sensors, such as temperature, humidity, light, etc. Next, the mean vector and covariance matrix of the historical environmental feature data set related to the current data point are extracted from the environmental feature database. These statistical data provide the normal fluctuation range of the data and the relationship between the features, which is the basis for anomaly detection. Using the current multidimensional feature vector and the mean vector and covariance matrix in the historical data, the system calculates the data anomaly detection value. The data anomaly detection value is compared with the preset first threshold in real time. If the data anomaly detection value is less than or equal to the first threshold, it indicates that the data is within the normal range, and the system will not perform any processing on the data. However, if the data anomaly detection value is greater than the first threshold, the system determines that the data point is abnormal data. Once the data is marked as abnormal, it will be eliminated from further analysis and processing. Doing so not only helps maintain the quality and reliability of data processing, but also prevents the risk of making misleading decisions based on erroneous data.

[0037] In this implementation scheme, by performing data anomaly detection and corresponding anomaly processing on the real-time environmental feature data set, the system significantly improves the accuracy and reliability of the data. By calculating the data anomaly detection value, the system can effectively identify and remove abnormal data points that deviate from the normal range. This not only prevents the risk of analysis and decision-making based on erroneous data, but also ensures the purity of the data set, thereby improving the accuracy of subsequent data processing and pest and disease monitoring. This step reduces the need for manual review and improves the overall efficiency and response speed of the system through automated anomaly identification and processing. In addition, timely removal of abnormal data from the analysis process helps the system to more accurately identify pest and disease patterns and trends, providing more reliable decision-making support for forestry management, thereby effectively improving the ability to protect forestry resources and prevent pest and disease losses.

[0038] Specifically, the specific calculation formula of the data anomaly detection value is: In the formula, AD represents the data anomaly detection value, x represents the multidimensional feature vector of the current data point, μ represents the mean vector, and ∑ represents the covariance matrix. This formula is a method to measure the distance between a point and a distribution, which is suitable for scenarios that consider the correlation between features and sensitivity to different scales. The mean vector μ and the covariance matrix ∑ are calculated from historical data in the environmental feature database. The mean vector is the average value of the historical data of each feature, reflecting the central trend of the data; while the covariance matrix describes the change relationship between the data features, indicating the linear dependence between the features and the variability of each feature data. Specifically, the mean vector μ is obtained by arithmetic averaging of the eigenvalues ​​of the historical data, while the covariance matrix ∑ is obtained by calculating the average value of the product of the deviations of each eigenvalue from its mean value. In the process of calculating the data anomaly detection value, it is first necessary to differentiate the multidimensional feature data vector x collected in real time from the mean vector μ, that is, to calculate the difference between x and μ. Then, the obtained difference vector is left-multiplied by the inverse matrix of the covariance matrix ∑, and then the dot product operation is performed with its own transposed vector, and finally the square root is taken to obtain the AD value.

[0039] In this embodiment, by implementing the above-mentioned data anomaly detection formula, data points that deviate from the normal pattern can be effectively identified and screened. This calculation process not only considers the correlation between the characteristics of the data points, but also integrates their relative scales and distributions, thereby providing a highly accurate anomaly detection method. The implementation of this method significantly improves the data quality of the monitoring system and ensures the reliability of the basic data for analysis and decision-making. By considering the overall statistical characteristics of the data, potential anomalies can be captured more sensitively, thereby timely eliminating erroneous data that may affect the accuracy of pest and disease monitoring results. As a result, the overall performance and responsiveness of the system are enhanced, ensuring the timeliness and accuracy of pest and disease warnings.

[0040] Specifically, interference detection is performed on the normal environmental feature data set processed by the abnormality detection module, and signal filtering is performed according to the interference detection result. The specific process is as follows: first, the sound feature vector is obtained from the data that has been verified to be normal. In addition, the system extracts historical pest and disease signal data and historical interference signal data from the environmental feature database, which have been marked and used for training and calibration processes. Next, the probability density value of the historical data is calculated, that is, the probability density value of the pest and disease signal and the probability density value of the interference signal. These probability density values ​​are obtained through a probabilistic statistical model, reflecting the frequency and pattern of occurrence of various types of signals in historical records. By combining the real-time collected sound feature vectors with these probability density values, the system can calculate a comprehensive interference data detection value. The real-time interference data detection value is compared with a preset second threshold. If the interference data detection value is less than or equal to the second threshold, the system determines that the current signal is an interference signal, which means that the signal does not match the known pest and disease pattern, so it is intercepted to prevent it from entering the next stage of the analysis process. On the contrary, when the interference data detection value is greater than the second threshold, the system determines that the sound feature vector belongs to the pest and disease signal data, and thus retains it for further analysis.

[0041] In this implementation scheme, by performing interference detection and signal filtering on the normal environmental feature data set processed by the abnormality inspection module, the system can effectively distinguish between real pest and disease signals and ordinary environmental interference. This process uses the probability density value of historical pest and disease and interference signal data, combined with the sound feature vector collected in real time, to calculate the interference data detection value. By comparing this value with the set threshold, the system can accurately intercept those signals that are considered to be interference to prevent them from misleading subsequent data analysis and decision-making processes. At the same time, this method ensures that all signals exceeding the threshold, that is, those data that may indicate pest and disease activity, are retained and used for further analysis. This precise filtering mechanism significantly improves the accuracy and efficiency of pest and disease monitoring, reduces the probability of false alarms, and makes the monitoring system more reliable and effective in a real-time environment, thereby greatly enhancing the timeliness and adaptability of forestry protection measures.

[0042] Specifically, the specific calculation formula of the interference data detection value is: Where, IDV represents the interference data detection value, which is used to distinguish whether the received signal is mainly caused by pests or other interference sources. P(S|Z0) represents the probability density value of historical pest signal data, which is used to distinguish whether the received signal is mainly caused by pests or other interference sources. P(S|Q1) is the probability density value of historical interference signal data, which is obtained based on historical interference data not related to pests and diseases. IDV provides a quantitative measure indicating that the current signal is more likely to be caused by pests and diseases rather than ordinary interference. When the IDV value is high, it indicates that P(S|Z0) is larger than P(S|Q1), that is, the signal is more likely to come from pests and diseases. Conversely, a lower IDV value indicates that the signal is more likely to be caused by interference.

[0043] In this embodiment, by applying the calculation formula of the interference data detection value, the system can effectively distinguish between signals caused by pests and diseases and signals generated by other environmental disturbances. This method provides a quantitative indicator to help determine the source of the currently monitored signal by comparing the probability density of historical pest and disease signals and interference signals. Higher interference data detection values ​​indicate that the signal is likely caused by pests and diseases, while lower values ​​suggest that the signal may originate from conventional environmental disturbances. This mechanism significantly improves the accuracy of the monitoring system's identification of pests and diseases and reduces the possibility of false alarms, making the early warning system more reliable. In addition, by ensuring that only high-confidence pest and disease signals are responded to, the system can allocate resources more effectively, optimize pest and disease management measures, and ensure that potential forest health issues are dealt with in a timely and accurate manner.

[0044] Specifically, pest and disease activity is monitored based on real-time data, and corresponding treatment measures are taken. The specific process is as follows: First, the pest and disease signal data processed by the anti-interference prediction processing module is received and analyzed in real time. This module ensures that the incoming data has been cleaned, and possible environmental noise and non-pest and disease related interference are eliminated. Only signals that are verified as possible pests and diseases will be further analyzed. Next, based on these screened pest and disease signal data, advanced analysis algorithms are used to detect whether there are abnormal pest and disease activities. This includes comparing current data with historical data patterns, using machine learning models to predict pest and disease development trends, and other methods to identify any abnormal or sudden pest and disease behavior. If the system detects abnormal pest and disease activity, the alarm system will be triggered immediately. This includes sending automated alarms to the communication equipment of forestry managers and related maintenance personnel to ensure a quick response.

[0045] In this implementation plan, by implementing real-time pest and disease activity monitoring and emergency response measures, the system greatly enhances the timeliness and accuracy of forest pest and disease management. This process uses advanced analytical techniques and algorithms to analyze processed pest and disease signal data in real time, and can quickly identify and respond to abnormal pest and disease activities. Once a potential pest and disease threat is detected, the system will automatically trigger an alarm and quickly notify forestry managers and related maintenance personnel through multiple communication channels to ensure that appropriate prevention and control measures can be taken in a timely manner. This real-time monitoring and rapid alarm capability not only reduces the ecological and economic losses that pests and diseases may cause, but also improves the effectiveness of prevention and control measures, ensuring the continued health and productivity of forestry resources. The implementation of this entire system has significantly improved the adaptability and response speed of forestry protection measures, and provided strong technical support for forestry pest and disease management.

[0046] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0047] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things is characterized by: include: The data acquisition module is used to rationally plan the location of sensors according to the forest terrain, vegetation type and historical pest and disease occurrence, and collect environmental characteristic data sets in real time; The preprocessing module is used to perform data vectorization, data cleaning, data synchronization and data standardization on the environmental feature data set collected in real time; The transmission module is used to collect and analyze the status of transmission nodes in real time, and select the best data transmission path for data transmission according to the node status analysis results; The anomaly detection module is used to perform data anomaly detection on the real-time environmental feature data set transmitted by the transmission module, and perform anomaly processing according to the results of the data anomaly detection, store the normal environmental feature data set, mark it as a historical environmental feature data set, obtain historical pest and disease signal data and historical interference signal data, and establish an environmental feature database; The anti-interference prediction processing module is used to perform interference detection on the normal environmental feature data set processed by the anomaly detection module, and perform signal filtering according to the interference detection results; The real-time early warning module is used to monitor pest and disease activities based on real-time data and take corresponding treatment measures.

2. According to claim 1, the intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things is characterized by: The specific process of real-time collection of environmental feature data sets is as follows: The environmental characteristic data set includes air temperature, air humidity, light intensity, vibration data, image data and sound signals; Use a temperature sensor to measure air temperature, use a temperature sensor to measure air humidity, use a light sensor to measure light intensity, use a vibration sensor to detect tiny vibrations of tree trunks and branches and leaves and mark them as vibration data, use a high-definition camera to obtain images of the plant surface and mark them as image data, and use a microphone to collect sounds such as insects' chirping and mark them as sound signals.

3. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things according to claim 1 is characterized by: The specific process of data vectorization, data cleaning, data synchronization and data standardization of the real-time collected environmental feature data set is as follows: Data vectorization: convert all collected environmental feature data sets into numerical vector form; organize air temperature, air humidity, light intensity, vibration data, image data and sound signals into multi-dimensional feature vectors, each of which represents a complete data record at a time point, marked as temperature feature vector, humidity feature vector, light intensity feature vector, vibration feature vector, image feature vector and sound feature vector respectively; Data cleaning: Detect and process missing values, outliers, and noise data in environmental feature datasets; Data synchronization: synchronize data from different sensors based on timestamps; Data standardization: Normalize the data in the environmental characteristic dataset to eliminate dimensional differences.

4. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things according to claim 1 is characterized by: The specific process of collecting and analyzing the status of transmission nodes in real time and selecting the best path for data transmission according to the node status analysis result is as follows: Node initialization: When each sensor node starts, it broadcasts its own status information, including remaining energy, location information, and environmental parameters; Neighbor discovery: Nodes build neighbor lists by receiving broadcast information from neighbor nodes to prepare for path selection; Path selection: obtain the number of hops in the transmission path, obtain the received signal strength, the remaining energy of the node, the node temperature, the node humidity and the node light intensity; calculate the optimal path selection value through the number of hops in the path, the received signal strength, the remaining energy of the node, the node temperature, the node humidity and the node light intensity, and according to the calculation results, select the path with the smallest optimal path selection value as the data transmission path.

5. According to claim 4, the intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things is characterized by: The specific calculation formula of the optimal path selection value is: In the formula, C path represents the optimal path selection value, n represents the number of hops in the path, HC i represents the number of hops in the i-th segment, w1 represents the weight coefficient of the number of hops in the i-th segment, RSSI i represents the received signal strength of the i-th segment, w2 represents the weight coefficient of the received signal strength of the i-th segment, E i represents the residual energy of the ith node, w3 represents the weight coefficient of the residual energy of the ith node, T i represents the node temperature of the i-th node, a represents the influence coefficient of the i-th node temperature, H i represents the node humidity of the i-th node, b represents the influence coefficient of the i-th node humidity, L i represents the node illumination intensity of the i-th node, c represents the influence coefficient of the i-th node illumination intensity, and w4 represents the weight coefficient of the environmental factor.

6. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things according to claim 1 is characterized by: The specific process of performing data anomaly detection on the real-time environmental feature data set transmitted by the transmission module and performing anomaly processing according to the result of data anomaly detection is as follows: Obtain the multidimensional feature vector of the current data point, obtain the mean vector and covariance matrix from the historical environmental feature data set of the environmental feature database, and obtain the data anomaly detection value by comprehensively calculating the multidimensional feature vector, mean vector and covariance matrix; The data anomaly detection value and the first threshold are compared in real time. When the data anomaly detection value is less than or equal to the first threshold, no processing is performed. When the data anomaly detection value is greater than the first threshold, it is determined to be abnormal data and the abnormal data is removed from the analysis.

7. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things according to claim 6 is characterized by: The specific calculation formula of the data anomaly detection value is: In the formula, AD represents the data anomaly detection value, x represents the multidimensional feature vector of the current data point, μ represents the mean vector, and ∑ represents the covariance matrix.

8. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things according to claim 1 is characterized by: The specific process of performing interference detection on the normal environmental feature data set processed by the anomaly detection module and filtering the signal according to the interference detection result is as follows: Acquire sound feature vectors, historical pest and disease signal data, and historical interference signal data, calculate the probability density value of the historical pest and disease signal data and the probability density value of the historical interference signal data, and obtain the interference data detection value through comprehensive calculation of the sound feature vectors, the probability density value of the historical pest and disease signal data, and the probability density value of the historical interference signal data; Compare the interference data detection value and the second threshold in real time. When the interference data detection value is less than or equal to the second threshold, it is determined to be an interference signal, and the data determined to be an interference signal is intercepted and does not enter the next link; When the interference data detection value is greater than the second threshold, it is determined to be pest signal data, and the data determined to be pest signal data is retained for subsequent analysis.

9. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things according to claim 1 is characterized by: The specific calculation formula of the interference data detection value is: Where IDV represents the interference data detection value, P(S|Z0) represents the probability density value of historical pest and disease signal data, and P(S|Q1) represents the probability density value of historical interference signal data.

10. The intelligent monitoring and early warning system for forestry pests and diseases based on the Internet of Things according to claim 1 is characterized by: The specific process of monitoring pest and disease activities based on real-time data and taking corresponding treatment measures is as follows: Receive and analyze pest and disease signal data processed by the anti-interference prediction processing module in real time, and detect whether there is abnormal pest and disease activity based on the real-time pest and disease signal data. If abnormal pest and disease activity is detected, trigger an alarm and notify relevant personnel.

Images