An intelligent collection method of integrated forest and grass data

Abstract

The application relates to the technical field of data collection, in particular to an intelligent collection method for integrated forest and grass data, which comprises the following steps: based on a target forest and grass area, deploying a UAV to the target area, capturing the image of the target forest and grass area through the camera loaded on the UAV, screening the captured image, eliminating the image with insufficient illumination and overexposure, and performing contrast optimization on the reserved image to obtain an optimized forest and grass image set. Through real-time image capturing and screening of the UAV in the target area, the application effectively eliminates the image with non-standard quality, makes the resource management decision more based on real-time and accurate data, adjusts the image collection strategy and data transmission frequency of the UAV, and according to the actual network state and the severity of the area, makes the data collection and transmission more efficient, reduces the waste of resources, and enhances the rapid response capability of the system in response to disasters.

Description

Technical Field

[0001] This invention relates to the field of data acquisition technology, and in particular to an intelligent acquisition method for integrating forestry and grassland data. Background Technology

[0002] The field of data acquisition technology involves the automated or manual collection of information using various hardware and software tools to process, analyze, and utilize data. Key components of this technology include sensors, data acquisition systems, and interface software. Data acquisition is crucial in many industries, particularly in environments requiring real-time monitoring and control, such as industrial automation, environmental monitoring, and healthcare. It not only helps improve the accuracy and reliability of data but also supports decision-making through real-time analysis, enhancing the predictability of process efficiency and outcomes.

[0003] The integrated intelligent data collection method for forestry and grassland ecosystems utilizes sensors and data analytics to monitor and collect data on various biotic and abiotic factors in forestry and grassland ecosystems in real time. Its main applications include, but are not limited to, forest health monitoring, sustainable management of forest products, disaster prevention, and ecological research. Through integrated data collection, the status of forestry and grassland resources can be assessed more accurately, supporting decision-making processes for environmental protection and sustainable resource utilization. Furthermore, this technology helps improve the efficiency and accuracy of forestry and grassland data management, providing support for scientific research and commercial applications.

[0004] Traditional data collection methods rely on standardized data acquisition processes, neglecting the preliminary screening and optimization of data quality. This leads to the processing of large amounts of low-quality data in practice, increasing the difficulty of analysis and the error rate. The lack of precise management of the spatial location of collected data makes it impossible to effectively identify and track the specific spread paths of pests and diseases. Traditional methods typically employ fixed data transmission strategies, failing to adjust according to actual network conditions and data importance. When the network is unstable or the data volume is large, this can result in data transmission delays or loss. These factors limit the efficiency of monitoring systems, reduce the timeliness and accuracy of pest and disease management, and affect the quality of decision-making for sustainable management of forestry and grassland resources and ecological protection. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an intelligent method for integrating forestry and grassland data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent data collection method for integrating forestry and grassland data, comprising the following steps:

[0007] S1: Based on the target forest and grassland area, deploy the drone to the target area, capture images of the target forest and grassland area through the camera mounted on the drone, filter the captured images to remove images with insufficient light and overexposure, and optimize the contrast of the retained images to obtain an optimized forest and grassland image set;

[0008] S2: Based on the optimized forest and grassland image set, feature extraction is performed on the vegetation in the images. The health status of the vegetation is analyzed according to the extracted features, the corresponding pest and disease types are identified, and the location of damaged plants and pests and diseases in the images is located. The coordinate data of each identification item is recorded to obtain the forest and grassland anomaly detection record.

[0009] S3: Based on the forest and grassland anomaly detection records, analyze the proportion of plants affected by pests and diseases in the region, assess the severity of pests and diseases based on the proportion of plants and the distribution density of the damaged areas, and use GIS technology to combine the data with geographic location information to construct a spatial distribution map of the affected areas and obtain spatial distribution information of pests and diseases.

[0010] S4: Based on the spatial distribution information of the pests and diseases, adjust the image acquisition strategy of the UAV, adjust the image acquisition frequency of the differentiated areas according to the severity of the differentiated areas, and adjust the frequency of data transmission back to the server according to the current network status to obtain a targeted forest and grassland monitoring dataset.

[0011] The present invention is improved in that the step of obtaining the optimized forest and grassland image set is as follows:

[0012] S111: Based on the target forest and grassland area, deploy the drone to the target area, and capture images of the target forest and grassland area through the camera mounted on the drone to obtain a basic image set;

[0013] S112: Based on the base image set, compare the image brightness with a preset brightness threshold, remove images with insufficient lighting and overexposure, and obtain the retained image set;

[0014] S113: Based on the retained image set, using the formula:

[0015] C = α C ×(C o -μ C )+β C +C o

[0016] Calculate the adjusted pixel value C to obtain the optimized forest and grassland image set, where C is the adjusted pixel value, and α is the value of the adjusted pixel value. C C is the contrast enhancement factor. o μ is the original pixel value in the image. C To preserve the average pixel value of the image set, β C This is the brightness adjustment factor.

[0017] The present invention is improved in that the method for identifying the corresponding pest and disease type is as follows:

[0018] S211: Based on the optimized forest and grassland image set, extract vegetation features from the images, including color and texture features, to obtain a vegetation feature set;

[0019] S212: Based on the aforementioned set of vegetation features, compare it with known pest and disease type features using the formula:

[0020]

[0021] Calculate the matching score M, where M is the matching score. f is the weight of the i-th feature. iD f is the i-th feature value of the image to be recognized. iP n represents the value of the corresponding feature in the pest and disease type database. M The total number of features.

[0022] S213: Based on the matching score M, compare it with the preset matching threshold, mark the pest and disease types with matching scores exceeding the preset matching threshold as matched, and obtain the pest and disease type identification result.

[0023] The present invention is improved in that the step of obtaining the forest and grassland anomaly detection record is as follows:

[0024] S221: Based on the pest and disease type identification results, locate the pixel coordinates of the damaged plants and pests in the image. Each damaged point and pest is identified by its pixel position to obtain pixel coordinate information.

[0025] S222: Based on the pixel coordinate information, using the formula:

[0026] (x g y g )=(s g ×x p +o x s g ×y p +o y )

[0027] Calculate the transformed map coordinates (x) g y g ), where (x p y p ) represents the pixel coordinates of the damaged plants and pests in the image, s g o is the scale transformation factor. x and o y It is the offset, (x) g yg () represents the converted map coordinates.

[0028] S223: Based on the transformed map coordinates (x) g y g By integrating the results of pest and disease type identification, abnormal forest and grassland detection records are obtained.

[0029] The present invention is improved in that the method for analyzing the proportion of plants affected by pests and diseases in the analyzed area is as follows:

[0030] S311: Based on the aforementioned forest and grassland anomaly detection records, extract the coordinates of plants affected by pests and diseases, analyze the total number of affected plants, and obtain plant quantity information;

[0031] S312: Based on the plant quantity information, and according to the total plant quantity in the area, using the formula:

[0032]

[0033] Calculate the percentage P of plants affected by pests and diseases to obtain the analysis results of the percentage of plants affected by pests and diseases, where N affected N represents the number of plants affected by pests and diseases in the region. total Let P be the total number of plants in the region, and let P be the percentage of plants affected by pests and diseases.

[0034] The present invention includes an improvement, wherein the method for assessing the severity of pests and diseases is as follows:

[0035] S321: Based on the analysis results of the proportion of plants affected by the pests and diseases, and by extracting the distribution density of the damaged areas, the severity correlation data is obtained;

[0036] S322: Based on the severity correlation data, using the formula:

[0037]

[0038] The severity index S of pests and diseases is calculated to obtain the assessment result of the severity of pests and diseases, where α S and β S γ is a weighting factor, where P is the proportion of plants affected by pests and diseases, D is the distribution density of the damaged area, and γ is the weighting factor. S is a normalization constant, and S is the severity index of pests and diseases.

[0039] The present invention is improved in that the method for adjusting the image acquisition frequency of the differentiated regions is as follows:

[0040] S411: Based on the spatial distribution information of pests and diseases, extract the severity index S of pests and diseases in the target area, and extract the basic image acquisition frequency to obtain acquisition frequency correlation data;

[0041] S412: Based on the aforementioned acquisition frequency correlation data, using the formula:

[0042]

[0043] Calculate the adjusted image acquisition frequency F adj The image acquisition frequency adjustment results were obtained, where S is the severity index of pests and diseases, and F... base Based on the image acquisition frequency, κ F F is the sensitivity adjustment constant. adj This refers to the adjusted image acquisition frequency.

[0044] The present invention is improved in that the steps for obtaining the targeted forestry and grassland monitoring dataset are as follows:

[0045] S421: Based on the image acquisition frequency adjustment result, extract the network latency and basic data transmission frequency according to the current network status to obtain data transmission association information;

[0046] S422: Based on the data transmission association information, using the formula:

[0047]

[0048] Calculate the adjusted data transmission frequency R adj , where L R Indicates network latency, η R It is the delay effect coefficient, R base Based on the basic data transmission frequency, ξ R R is the normalization constant. adj The adjusted data transmission frequency is given by e, where e is the base of the natural logarithm.

[0049] S423: Based on the adjusted data transmission frequency R adj Data transmission was performed to obtain a targeted forestry and grassland monitoring dataset.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0051] In this invention, real-time image capture and filtering by drones in the target area effectively eliminates images that do not meet quality standards, and the contrast of the retained images is optimized, improving the accuracy and usability of the data. In particular, by extracting features from the images and analyzing their health status, the specific locations of damaged plants and pests are located, enhancing the detail of the data and making forest area management more precise. The spatial distribution map of pests and diseases constructed using GIS technology provides a dynamic and intuitive method for assessing the status of forest and grassland resources, enabling resource management decisions to be more based on real-time and accurate data. The image acquisition strategy and data transmission frequency of the drones are adjusted according to the actual network status and the severity of the area, making data collection and transmission more efficient, reducing resource waste, and enhancing the system's rapid response capability to disasters. Attached Figure Description

[0052] Figure 1 This is a flowchart of the method of the present invention;

[0053] Figure 2 This is a flowchart illustrating the process of obtaining an optimized forest and grassland image set according to the present invention;

[0054] Figure 3 This is a flowchart illustrating the identification of corresponding pest and disease types in this invention;

[0055] Figure 4 This is a flowchart illustrating the process of obtaining forest and grassland anomaly detection records according to the present invention;

[0056] Figure 5 This is a flowchart illustrating the percentage of plants affected by pests and diseases within a given area, as described in this invention.

[0057] Figure 6 This is a flowchart illustrating the assessment of the severity of pests and diseases according to the present invention.

[0058] Figure 7 This is a flowchart illustrating the process of adjusting the image acquisition frequency in differentiated regions according to the present invention.

[0059] Figure 8 This is a flowchart illustrating the process of obtaining a targeted forestry and grassland monitoring dataset according to the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0061] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0062] Please see Figure 1 This invention provides a technical solution: an intelligent data collection method for integrating forestry and grassland data, comprising the following steps:

[0063] S1: Based on the target forest and grassland area, deploy the drone to the target area, capture images of the target forest and grassland area through the camera mounted on the drone, filter the captured images to remove images with insufficient light and overexposure, and optimize the contrast of the retained images to obtain an optimized forest and grassland image set;

[0064] S2: Based on the optimized forest and grassland image set, feature extraction is performed on the vegetation in the image. The health status of the vegetation is analyzed according to the extracted features, the corresponding pest and disease types are identified, the location of damaged plants and pests in the image is located, the coordinate data of each identification item is recorded, and the forest and grassland anomaly detection record is obtained.

[0065] S3: Based on the abnormal detection records of forests and grasslands, analyze the proportion of plants affected by pests and diseases in the region. Based on the proportion of plants and the distribution density of the damaged areas, assess the severity of pests and diseases. Use GIS technology to combine the data with geographic location information to construct a spatial distribution map of the affected areas and obtain spatial distribution information of pests and diseases.

[0066] S4: Based on the spatial distribution information of pests and diseases, adjust the image acquisition strategy of UAVs, adjust the image acquisition frequency of differentiated areas according to the severity of differentiated areas, and adjust the data transmission frequency back to the server according to the current network status to obtain a targeted forestry and grassland monitoring dataset.

[0067] The optimized forest and grassland image set includes filtered images and images with contrast optimization. The forest and grassland anomaly detection record includes the location coordinates of damaged trees, pest and disease type identification, and affected area data. The spatial distribution information of pests and diseases includes pest and disease density information, the affected area range of the damaged area, and the distribution information of pest and disease types. The targeted forest and grassland monitoring dataset includes the updated image acquisition frequency, real-time acquired image data, and adjusted data transmission frequency.

[0068] Please see Figure 2 The steps for optimizing the acquisition of forest and grassland image sets are as follows:

[0069] S111: Based on the target forest and grassland area, deploy the drone to the target area, and capture images of the target forest and grassland area through the camera mounted on the drone to obtain a basic image set;

[0070] S112: Based on the base image set, compare the image brightness with the preset brightness threshold, remove images with insufficient light and overexposed images, and obtain the retained image set;

[0071] S113: Based on the retained image set, using the formula:

[0072] C = α C ×(C o -μ C )+β C +C o

[0073] Calculate the adjusted pixel value c to obtain the optimized forest and grassland image set, where c is the adjusted pixel value, and α is the value of the adjusted pixel value. C C is the contrast enhancement factor. o μ is the original pixel value in the image. C To preserve the average pixel value of the image set, β C This is the brightness adjustment factor.

[0074] formula:

[0075] C = α C ×(C o -μ C )+β C +C o

[0076] The meaning and acquisition method of the parameters:

[0077] α C Contrast enhancement factor: Used to control the stretching or compression of image contrast. It is usually set manually or obtained through experimental data according to the specific needs of the image. The specific value can be adjusted by the user's visual feedback to achieve the best effect.

[0078] C o The raw pixel values ​​in the image are read directly from the image data.

[0079] μ C The average pixel value of the retained image set is calculated by taking the arithmetic mean of all pixel values ​​in the selected retained image set.

[0080] β C Brightness adjustment factor: Used to adjust the overall brightness of the image to ensure that the image brightness is within a reasonable range. This parameter is also set based on experiments or user feedback, and the common value range may be from -20 to 20.

[0081] Calculation Example

[0082] Set the pixel value C in the original image o =120, the average pixel value μ of the selected image set. C =100, contrast enhancement factor α C =1.2, brightness adjustment coefficient β C =15.

[0083] Calculate C o -μ C 120 - 100 = 20

[0084] Multiply the result by the contrast enhancement factor α C 1.2 × 20 = 24

[0085] Add brightness adjustment factor β C and the original pixel value C o 24 + 15 + 120 = 159

[0086] The optimized pixel value C=159, which is an improvement over the original pixel value of 120. This enhances the image contrast and adjusts the brightness appropriately, making the image details more obvious and the overall visual effect clearer.

[0087] Please see Figure 3 The method for identifying the corresponding pest and disease type is as follows:

[0088] S211: Based on the optimized forest and grassland image set, extract vegetation features from the images, including color and texture features, to obtain a vegetation feature set;

[0089] S212: Based on the set of vegetation features, compare them with known pest and disease type features using the formula:

[0090]

[0091] Calculate the matching score M, where M is the matching score. f is the weight of the i-th feature. iD f is the i-th feature value of the image to be recognized. iP n represents the value of the corresponding feature in the pest and disease type database. M The total number of features.

[0092] S213: Based on the matching score M, compare it with the preset matching threshold, mark the pest and disease types with matching scores exceeding the preset matching threshold as matches, and obtain the pest and disease type identification results.

[0093] formula:

[0094]

[0095] Parameter details and acquisition methods:

[0096] n M : Represents the total number of features, which is determined based on the actual number of features in the feature set of the image to be analyzed. For example, an image may have 10 features such as color and texture.

[0097] The weight of the i-th feature quantifies the importance of each feature. It is usually set by domain experts based on the influence of the feature in pest and disease identification. For example, the importance score of a feature may be between 0.1 and 0.9.

[0098] f iD and f iP :f iD It is the i-th feature value of the image to be identified, obtained directly from the image feature extraction process; f iP It is the i-th value of the corresponding feature in the database of known pest and disease types, which is usually pre-calculated and stored from historical datasets.

[0099] Calculation example:

[0100] Set the following parameters:

[0101] Feature number n M =3 (e.g., color histogram, texture histogram, edge histogram), weight Feature value f of the image to be identified 1D =0.8, f 2D =0.3, f 3D =0.5 Known pest and disease characteristic value f 1P =0.7, f 2P =0.5, f 3P =0.6.

[0102] The calculation steps are as follows:

[0103] Calculate the matching score for the first feature:

[0104]

[0105] Calculate the matching score for the second feature:

[0106]

[0107] Calculate the matching score for the third feature:

[0108]

[0109] The weighted sum of all feature matching scores:

[0110] M=M1+M2+M3=0.27+0.4+0.18=0.85

[0111] The result M=0.85 indicates that the image has a high similarity to known pest and disease types, suggesting that the image is likely affected by this type of pest and disease.

[0112] Please see Figure 4 The steps for obtaining forest and grassland anomaly detection records are as follows:

[0113] S221: Based on the results of pest and disease type identification, locate the pixel coordinates of damaged plants and pests in the image. Each damaged point and pest is identified by its pixel position, and pixel coordinate information is obtained.

[0114] S222: Based on pixel coordinate information, using the formula:

[0115] (x g y g )=(s g ×x p +o x s g ×y p +o y )

[0116] Calculate the transformed map coordinates (x) g y g ), where (x p y p ) represents the pixel coordinates of the damaged plants and pests in the image, s g o is the scale transformation factor. x and o y It is the offset, (x) g y g () represents the converted map coordinates.

[0117] S223: Based on the transformed map coordinates (x g y g By integrating the results of pest and disease type identification, abnormal forest and grassland detection records are obtained.

[0118] formula:

[0119] (x g y g )=(s g ×x p +o x s g ×y p +o y )

[0120] Parameter details and acquisition methods:

[0121] s g : Scale transformation factor, used to convert the image coordinate system to the map coordinate system. This factor is obtained through field measurements and calculations based on the resolution differences between the map and the image. For example, if 1 pixel equals 0.5 meters, then s g =0.5.

[0122] o x and o y Offset: This represents the horizontal and vertical deviation between the image coordinate system and the map coordinate system. The parameter is typically obtained by marking known geographical locations on the image.

[0123] x p y p The coordinates of the points in the image to be converted are obtained directly from the image processing algorithm and represent the location of the identified damaged plants or pests in the image.

[0124] Calculation example:

[0125] The pixel coordinates of the damaged plants identified in the image are set to x. p =200, y p =300, scale conversion factor

[0126] s g =0.5 (i.e., 0.5 meters per pixel), offset o x =10, o y =15.

[0127] The calculation steps are as follows:

[0128] x-coordinates of the image p Convert to actual map coordinates:

[0129] x g =s g ×x p +o x =0.5×200+10=100+10=110

[0130] The pixel coordinates y of the image p Convert to actual map coordinates:

[0131] y g =s g ×y p +o y =0.5×300+15=150+155165

[0132] The obtained coordinates (x) g y gThe coordinates (110, 165) represent the location of the damaged plant on the map. These coordinates help Geographic Information Systems (GIS) or field personnel accurately pinpoint the affected area, enabling appropriate management or treatment measures. This method accurately maps each anomaly identified in the image to its specific location on the ground, effectively assisting in forestry and grassland management.

[0133] Please see Figure 5 The method for analyzing the proportion of plants affected by pests and diseases in the region is as follows:

[0134] S311: Based on the forest and grassland anomaly detection records, extract the coordinates of plants affected by pests and diseases, analyze the total number of affected plants, and obtain plant quantity information;

[0135] S312: Based on plant quantity information, according to the total number of plants in the area, using the formula:

[0136]

[0137] Calculate the percentage P of plants affected by pests and diseases to obtain the analysis results of the percentage of plants affected by pests and diseases, where N affected N represents the number of plants affected by pests and diseases in the region. total Let P be the total number of plants in the region, and let P be the percentage of plants affected by pests and diseases.

[0138] formula:

[0139]

[0140] Parameter details and acquisition methods:

[0141] N effected This indicates the total number of plants affected by pests and diseases within the region. This number is derived from statistics of plants marked as damaged in forestry and grassland anomaly detection records.

[0142] N total Total number of plants in the area, obtained through statistics of all plants in the area using a Geographic Information System (GIS) or through field surveys.

[0143] Calculation example:

[0144] Within the target survey area, the number N of plants affected by pests and diseases is counted based on forest and grassland anomaly detection records. affected =150. Based on GIS technology or field surveys, the total number of plants N in this area is... total =1000.

[0145] Calculate the percentage of plants affected by pests and diseases:

[0146]

[0147] A calculation result of P = 15% indicates that 15% of the plants in the area are affected by pests and diseases. This percentage can help decision-makers understand the distribution and potential impact of pests and diseases, thereby enabling them to develop more effective control and management measures.

[0148] Please see Figure 6 The method for assessing the severity of pests and diseases is as follows:

[0149] S321: Based on the analysis results of the proportion of plants affected by pests and diseases, and by extracting the distribution density of the damaged areas, severity correlation data are obtained;

[0150] S322: Based on severity correlation data, using the formula:

[0151]

[0152] The severity index S of pests and diseases is calculated to obtain the assessment result of the severity of pests and diseases, where α S and β S γ is a weighting factor, where P is the proportion of plants affected by pests and diseases, D is the distribution density of the damaged area, and γ is the weighting factor. S is a normalization constant, and S is the severity index of pests and diseases.

[0153] formula:

[0154]

[0155] Parameter details and acquisition methods:

[0156] P: The percentage of plants affected by pests and diseases, calculated through previous steps.

[0157] D: Distribution density of damaged areas, usually obtained from GIS analysis, indicates the degree of concentration of damaged plants in geographic space.

[0158] α S Weighting factors enhance the influence of plant proportion in the assessment and are set based on historical data and expert experience.

[0159] β S The weighting factor adjusts the contribution of the distribution density to the evaluation results and is also set based on expert opinions and previous research results.

[0160] γ S : Normalization constant, used to ensure the rationality and stability of the formula results, and is usually adjusted according to the range and distribution of the data.

[0161] Calculation example;

[0162] In the target area, the percentage of plants affected by pests and diseases is set to P = 25% (i.e., 0.25), the distribution density of the damaged area is set to D = 50 (unit: damaged plants / square kilometer), and a weighting factor a is set. S =0.6, β S =0.4, normalization constant γ S =2.0.

[0163] The calculation steps are as follows:

[0164] Calculate P 2 :

[0165] P 2 =(0.25) 2 =0.0625

[0166] calculate

[0167]

[0168] Calculate S:

[0169]

[0170] The resulting pest and disease severity index, S≈1.43, is relatively low, indicating that the pest and disease severity in the area is mild. This calculation method allows for more flexible consideration of various influencing factors and can be adapted to different regions and pest and disease types by adjusting weighting factors and normalization constants.

[0171] Please see Figure 7 The method for adjusting the image acquisition frequency in differentiated regions is as follows:

[0172] S411: Based on the spatial distribution information of pests and diseases, extract the severity index S of pests and diseases in the target area, and extract the basic image acquisition frequency to obtain acquisition frequency correlation data;

[0173] S412: Based on the data associated with the acquisition frequency, using the formula:

[0174]

[0175] Calculate the adjusted image acquisition frequency F adj The image acquisition frequency adjustment results were obtained, where S is the severity index of pests and diseases, and F... base Based on the image acquisition frequency, κ F F is the sensitivity adjustment constant. adj This refers to the adjusted image acquisition frequency.

[0176] formula:

[0177]

[0178] Parameter details and acquisition methods

[0179] F base The base image acquisition frequency is the standard operating frequency of the drone before any adjustments are made. This frequency is determined based on the drone's performance and battery life, and is typically set to once per hour.

[0180] S: Severity index of pests and diseases, calculated through previous steps.

[0181] κ F Sensitivity adjustment constant: This constant controls how the pest and disease index affects the adjustment of the sampling frequency. This constant depends on the region's monitoring needs and the operational capabilities of the drones.

[0182] Calculation Example

[0183] Set the basic image acquisition frequency F base = 1 time / hour, pest and disease severity index S = 10, adjustment constant κ F =10.

[0184] The calculation steps are as follows:

[0185] Calculate the logarithmic term: log(S+1) = log(10+1) = log(11) ≈ 1.04

[0186] Calculate F adj :

[0187]

[0188] The adjusted image acquisition frequency is approximately 1.104 times per hour. This means that in cases of severe pest and disease infestation, the number of drone flights will increase slightly to ensure sufficient monitoring and data collection in high-risk areas.

[0189] Please see Figure 8 The steps for obtaining the targeted forestry and grassland monitoring dataset are as follows:

[0190] S421: Based on the image acquisition frequency adjustment results, extract the network latency and basic data transmission frequency according to the current network status to obtain data transmission association information;

[0191] S422: Based on data transmission association information, using the formula:

[0192]

[0193] Calculate the adjusted data transmission frequency R adj , where L R Indicates network latency, η R It is the delay effect coefficient, Rbase Based on the basic data transmission frequency, ξ R R is the normalization constant. adj The adjusted data transmission frequency is given by e, where e is the base of the natural logarithm.

[0194] S423: Based on the adjusted data transmission frequency R adj Data transmission was performed to obtain a targeted forestry and grassland monitoring dataset.

[0195] formula:

[0196]

[0197] Parameter details and acquisition methods:

[0198] R base The basic data transmission frequency is the standard operating frequency of the drone when network conditions are ideal. It is usually set according to the default settings of the drone and network equipment, such as once every 10 minutes.

[0199] L R Network latency, usually measured in milliseconds, can be obtained in real time using network monitoring tools. It represents the data transmission delay from the drone to the server.

[0200] η R Delay impact factor: This is an adjustment parameter used to define the strength of the impact of delay on transmission frequency. It is usually preset based on equipment performance and network quality.

[0201] ξ R : Normalization constant, used to stabilize the frequency change after adjustment, ensuring that the transmission frequency remains within a reasonable range under extreme network conditions.

[0202] Calculation Examples and Procedures

[0203] Set the following parameter values ​​for calculation: R base = 6 times / hour (once every 10 minutes), L R = 100ms (network latency), η R =0.05 (delay effect coefficient), ξ R =10 (normalization constant).

[0204] Calculate the exponential impact of network latency:

[0205]

[0206] Calculate R adj :

[0207]

[0208] Calculations show that, taking into account a network latency of 100ms, the adjusted transmission frequency increases slightly to approximately 6.004 times per hour.

[0209] Adjusted frequency R adj The frequency of approximately 6.004 times per hour indicates that even with slight network latency, the data transmission frequency remains almost unaffected, staying close to the fundamental frequency and ensuring efficient and stable data transmission. This method allows the UAV's data transmission strategy to flexibly adapt to different network environments, ensuring the timely transmission of critical monitoring data.

[0210] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. 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 scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for intelligent collection of integrated forestry and grassland data, characterized in that, Includes the following steps: Based on the target forest and grassland area, drones are deployed to the target area, and images of the target forest and grassland area are captured by the cameras mounted on the drones. By filtering the captured images, images with insufficient light and overexposure are removed, and the contrast of the retained images is optimized to obtain an optimized forest and grassland image set. Based on the optimized forest and grassland image set, feature extraction is performed on the vegetation in the images. The health status of the vegetation is analyzed based on the extracted features, the corresponding pest and disease types are identified, and the location of damaged plants and pests in the images is located. The coordinate data of each identified item is recorded to obtain the forest and grassland anomaly detection record. Based on the aforementioned forest and grassland anomaly detection records, the proportion of plants affected by pests and diseases in the region is analyzed. Based on the proportion of plants and the distribution density of the damaged areas, the severity of pests and diseases is assessed. GIS technology is used to combine the data with geographic location information to construct a spatial distribution map of the affected areas and obtain spatial distribution information of pests and diseases. Based on the spatial distribution information of pests and diseases, the image acquisition strategy of the UAV is adjusted. The image acquisition frequency of the differentiated areas is adjusted according to the severity of the differentiated areas, and the frequency of data transmission back to the server is adjusted according to the current network status to obtain a targeted forestry and grassland monitoring dataset.

2. The intelligent data acquisition method for integrating forestry and grassland data according to claim 1, characterized in that, The steps for obtaining the optimized forest and grassland image set are as follows: Based on the target forest and grassland area, drones are deployed to the target area, and images of the target forest and grassland area are captured by the cameras mounted on the drones to obtain a basic image set; Based on the base image set, the image brightness is compared with a preset brightness threshold, and images with insufficient lighting and overexposure are removed to obtain the retained image set. Based on the retained image set, using the formula: C=a C ×(C o -m C )+b C +C o Calculate the adjusted pixel value C to obtain the optimized forest and grassland image set, where C is the adjusted pixel value, and α is the value of the adjusted pixel value. C C is the contrast enhancement factor. o μ is the original pixel value in the image. C To preserve the average pixel value of the image set, β C This is the brightness adjustment factor.

3. The intelligent data acquisition method for integrating forestry and grassland data according to claim 1, characterized in that, The method for identifying the corresponding pest and disease type is as follows: Based on the optimized forest and grassland image set, vegetation features, including color and texture features, are extracted from the images to obtain a vegetation feature set; Based on the aforementioned set of vegetation features, a comparison is made with known pest and disease types using the formula: Calculate the matching score M, where M is the matching score. f is the weight of the i-th feature. iD f is the i-th feature value of the image to be recognized. P These are the values ​​of the corresponding features in the pest and disease type database. n M The total number of features. Based on the matching score M, it is compared with a preset matching threshold. Pests and diseases with matching scores exceeding the preset matching threshold are marked as matches, thus obtaining the pests and diseases type identification result.

4. The intelligent data acquisition method for integrating forestry and grassland data according to claim 3, characterized in that, The steps for obtaining the forest and grassland anomaly detection records are as follows: Based on the pest and disease type identification results, the pixel coordinates of the damaged plants and pests in the image are located. Each damaged point and pest is identified by its pixel position, and pixel coordinate information is obtained. Based on the pixel coordinate information, using the formula: (x g ,y g )=(s g ×x p +o x ,s g ×y p +o y ) Calculate the transformed map coordinates (x) g y g ), where (x p y p ) represents the pixel coordinates of the damaged plants and pests in the image, s g o is the scale transformation factor. x and o y It is the offset, (x) g y g () represents the converted map coordinates. Based on the transformed map coordinates (x) g y g By integrating the results of pest and disease type identification, abnormal forest and grassland detection records are obtained.

5. The intelligent data acquisition method for integrating forestry and grassland data according to claim 1, characterized in that, The method for analyzing the percentage of plants affected by pests and diseases within the analyzed area is as follows: Based on the aforementioned forest and grassland anomaly detection records, the coordinates of plants affected by pests and diseases are extracted, the total number of affected plants is analyzed, and plant quantity information is obtained. Based on the plant quantity information, and according to the total plant quantity in the region, the formula is used: Calculate the percentage P of plants affected by pests and diseases to obtain the analysis results of the percentage of plants affected by pests and diseases, where N affected N represents the number of plants affected by pests and diseases in the region. total Let P be the total number of plants in the region, and let P be the percentage of plants affected by pests and diseases.

6. The intelligent data acquisition method for integrating forestry and grassland data according to claim 5, characterized in that, The method for assessing the severity of pests and diseases is as follows: Based on the analysis results of the proportion of plants affected by the pests and diseases, and by extracting the distribution density of the damaged areas, severity correlation data were obtained. Based on the severity correlation data, using the formula: The severity index S of pests and diseases is calculated to obtain the assessment result of the severity of pests and diseases, where α S and β S γs is the weighting factor, P is the proportion of plants affected by pests and diseases, D is the distribution density of the damaged area, γs is the normalization constant, and S is the severity index of pests and diseases.

7. The intelligent data acquisition method for integrating forestry and grassland data according to claim 1, characterized in that, The method for adjusting the image acquisition frequency of the differentiated regions is as follows: Based on the spatial distribution information of pests and diseases, the severity index S of pests and diseases in the target area is extracted, and the basic image acquisition frequency is extracted to obtain acquisition frequency correlation data. Based on the aforementioned frequency-related data, using the formula: Calculate the adjusted image acquisition frequency F adj The image acquisition frequency adjustment results were obtained, where S is the severity index of pests and diseases, and F... base Based on the image acquisition frequency, κ F F is the sensitivity adjustment constant. adj This refers to the adjusted image acquisition frequency.

8. The intelligent data acquisition method for integrating forestry and grassland data according to claim 7, characterized in that, The steps for obtaining the targeted forestry and grassland monitoring dataset are as follows: Based on the image acquisition frequency adjustment results, and according to the current network status, network latency and basic data transmission frequency are extracted to obtain data transmission association information; Based on the data transmission association information, using the formula: Calculate the adjusted data transmission frequency R adj , where L R Indicates network latency, η R It is the delay effect coefficient, R base Based on the basic data transmission frequency, ξ R R is the normalization constant. adj The adjusted data transmission frequency, e is the base of the natural logarithm; Based on the adjusted data transmission frequency R adj Data transmission was performed to obtain a targeted forestry and grassland monitoring dataset.

Images