A forest resource asset determination method and system

Abstract

The application discloses a kind of forest resource assets determination method and system, the technical scheme of the present application is segmented to target forest area range Remote Sensing image acquisition is carried out, after calculating the resource value corresponding to each resource feature, sequentially arranged all target Remote Sensing image is sequentially fused, to accurately calculate the total value of forest resource assets of target forest area range;With the inaccuracy of the measurement between image acquisition and long continuous image in the prior art, the technical problem that the strategy of calculating part of forest interval using big data can only be calculated in a small range, and cannot be accurately calculated in the face of large area and wide range of forest range;And for large area and wide range of forest range, the actual situation of forest resource assets can be effectively, quickly and accurately determined, to provide strong data support for ecological environment protection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data processing, and particularly relates to a forest resource asset determination method and system. BACKGROUND

[0002] The determination of forest resource assets provides strong data support for ecological environment protection. However, the traditional way of detecting forest resource assets is to calculate by manpower estimation such as manual field detection, drawing of maps, and resource stock estimation, which cannot effectively, quickly and accurately determine the actual situation of forest resource assets.

[0003] With the increasing voice of ecological improvement and the increasing pace of ecological environment protection, how to get rid of the traditional manual calculation of forest resource assets and effectively, quickly and accurately determine the forest resource assets by using big data technology has become a technical problem to be solved in the market.

[0004] Although there are strategies for using big data to calculate resources in some forest areas on the market, due to the inaccuracy of image acquisition and long continuous image calculation, the existing technology for calculating resources in some forest areas by using big data can only calculate in a small range, and cannot accurately calculate in a large area and wide range of forest. SUMMARY

[0005] The present application provides a forest resource asset determination method and system, which can effectively, quickly and accurately determine the actual situation of forest resource assets for a large area and wide range of forest, and provide strong data support for ecological environment protection.

[0006] To solve the above technical problems, the present application provides a forest resource asset determination method, comprising:

[0007] Obtain the target latitude and longitude information of the forest resource assets to be detected, and perform segmented remote sensing image acquisition on the target forest area range in the target latitude and longitude information by using a UAV to obtain a plurality of target remote sensing images;

[0008] Respectively identify the features of each target remote sensing image, mark the resource features in the target remote sensing image, and obtain the corresponding resource features in each target remote sensing image;

[0009] Calculate the area size of each resource feature in the target remote sensing image where it is located to obtain an area value, and calculate the resource value corresponding to each resource feature according to the area value;

[0010] arranging all the target remote sensing images in order according to the target latitude and longitude information, and determining a reference resource feature point according to a resource feature marked in each of the target remote sensing images, and sequentially fusing all the target remote sensing images arranged in order according to the reference resource feature point to obtain a long continuous remote sensing image;

[0011] calculating a total value of forest resources assets of the target forest region range according to a resource value corresponding to each of the resource features in the long continuous remote sensing image.

[0012] As a preferred solution, the step of collecting segmented remote sensing images of the target forest region range in the target latitude and longitude information by the unmanned aerial vehicle to obtain multiple target remote sensing images specifically includes:

[0013] determining a maximum collection range according to the model of the unmanned aerial vehicle, and dividing the target latitude and longitude information into multiple latitude and longitude ranges according to the maximum collection range and arranging them in order;

[0014] controlling the unmanned aerial vehicle to collect segmented remote sensing images according to the arranged latitude and longitude ranges and to transmit the collected target remote sensing images in real time;

[0015] performing resource boundary feature recognition on the target remote sensing images transmitted back in real time, and determining the resource boundary as an end point of the target remote sensing image of the current segment when it is determined that the distance value between the resource boundary and the next target remote sensing image is less than a preset distance value;

[0016] re-dividing the latitude and longitude ranges that have not been subjected to image collection according to the maximum collection range and arranging them in order until the end points of the target remote sensing images of each segment are determined.

[0017] As a preferred solution, the step of performing feature recognition on each of the target remote sensing images, marking resource features in the target remote sensing images, and obtaining corresponding resource features in each of the target remote sensing images specifically includes:

[0018] transmitting each of the target remote sensing images as an input image to a preset resource feature recognition model for resource feature recognition to obtain an output image, wherein the output image marks the resource features in the target remote sensing images;

[0019] The preset resource feature recognition model is used to recognize resource features in an input image and determine a resource type, and outputs an output image marked with resource features and corresponding resource types.

[0020] As a preferred solution, the construction process of the preset resource feature recognition model specifically includes:

[0021] acquire a historical remote sensing image, mark forest resource features in the historical remote sensing image, and generate first marking information;

[0022] determine a corresponding resource type according to the marked forest resource features, generate second marking information, and mark the first marking information and the second marking information in the historical remote sensing image after association, and generate a final remote sensing image;

[0023] construct an initial identification model through a neural network algorithm, input the final remote sensing image into the initial identification model for training, and obtain a trained identification model after the number of training reaches a threshold number of times;

[0024] input the historical remote sensing image into the trained identification model for testing, and generate a resource feature identification model when a test success rate reaches a threshold success rate.

[0025] As a preferred solution, in the step of calculating the area size of each resource feature in the target remote sensing image where the resource feature is located to obtain an area value, and calculating a resource value corresponding to each resource feature according to the area value, the step specifically comprises:

[0026] determining a center point of each resource feature, determining a circumscribed circle of the resource feature with the center point as the center, and determining an area value corresponding to the resource feature according to the area of the circumscribed circle;

[0027] determining a target resource type of the resource feature corresponding to the circumscribed circle according to the output of the preset resource feature identification model marking the resource feature and the corresponding resource type;

[0028] determining a preset weight value according to the target resource type, calculating the product of the area value and the preset weight value as the resource value corresponding to the resource feature.

[0029] As a preferred solution, in the step of determining a reference resource feature point according to the resource features marked in each section of the target remote sensing image, and sequentially fusing all target remote sensing images arranged in sequence according to the reference resource feature point to obtain a long continuous remote sensing image, the step specifically comprises:

[0030] determining a terminal point of the current section of the target remote sensing image as a first fusion point in the resource features marked in each section of the target remote sensing image;

[0031] determining a resource boundary corresponding to the terminal point of the current section of the target remote sensing image as a second fusion point in the next section of the adjacent target remote sensing image;

[0032] The first fusion point and the second fusion point are aligned as reference resource feature points, and a target remote sensing image of a current segment and a target remote sensing image of a next segment are fused, until all target remote sensing images are sequentially fused to obtain a long continuous remote sensing image.

[0033] As a preferred scheme, the calculation formula of the total value of forest resource assets is:

[0034]

[0035] Wherein, S0 is the total value of forest resource assets; S i is the resource value corresponding to the i th resource feature, k i is a preset type threshold corresponding to the resource type of the i th resource feature; i is a resource feature, and n is the total number of resource features.

[0036] Correspondingly, the present application also provides a forest resource asset determination system, comprising: an image acquisition module, a feature marking module, an area determination module, an image fusion module and a total value calculation module.

[0037] The image acquisition module is used for acquiring target latitude and longitude information of forest resource assets to be detected, and acquiring a plurality of target remote sensing images by segmenting remote sensing image acquisition of a target forest area range in the target latitude and longitude information through a UAV.

[0038] The feature marking module is used for respectively marking resource features in each target remote sensing image, and obtaining corresponding resource features in each target remote sensing image.

[0039] The area determination module is used for calculating the area size of each resource feature in the target remote sensing image where the resource feature is located to obtain an area value, and calculating a resource value corresponding to each resource feature according to the area value.

[0040] The image fusion module is used for arranging all target remote sensing images in order according to the target latitude and longitude information, and determining reference resource feature points according to the resource features marked in each target remote sensing image, and sequentially fusing all target remote sensing images arranged in order according to the reference resource feature points to obtain a long continuous remote sensing image.

[0041] The total value calculation module is used for calculating the total value of forest resource assets of the target forest area range according to the resource value corresponding to each resource feature in the long continuous remote sensing image.

[0042] As a preferred solution, the image acquisition module is specifically configured to: determine a maximum acquisition range according to the model of the unmanned aerial vehicle, divide the target latitude and longitude information into multiple latitude and longitude ranges according to the maximum acquisition range and arrange them in sequence; control the unmanned aerial vehicle to sequentially perform segmented remote sensing image acquisition according to the arranged latitude and longitude ranges, and transmit the acquired target remote sensing images in real time; perform resource boundary feature recognition on the target remote sensing images transmitted back in real time, and when it is determined that the distance value between the identified resource boundary and the next segment of the target remote sensing image is less than a preset distance value, determine the resource boundary as the end point of the current segment of the target remote sensing image; re-divide the latitude and longitude ranges that have not been subjected to image acquisition according to the maximum acquisition range and arrange them in sequence until the end point of each segment of the target remote sensing image is determined.

[0043] As a preferred solution, the feature marking module is specifically configured to: respectively transmit each segment of the target remote sensing image as an input image to a preset resource feature recognition model for resource feature recognition, to obtain an output image, wherein the resource features in the target remote sensing image are marked in the output image; and wherein the preset resource feature recognition model is configured to identify resource features in an input image and determine a resource type, and output an output image in which resource features and corresponding resource types are marked.

[0044] As a preferred solution, the construction process of the preset resource feature recognition model specifically includes: obtaining historical remote sensing images, marking forest resource features in the historical remote sensing images to generate first marking information; determining corresponding resource types according to the marked forest resource features to generate second marking information, and associating the first marking information and the second marking information and marking them in the historical remote sensing images to generate final remote sensing images; constructing an initial recognition model through a neural network algorithm, inputting the final remote sensing images into the initial recognition model for training until the number of training reaches a number threshold, to obtain a trained recognition model; inputting the historical remote sensing images into the trained recognition model for testing until the test success rate reaches a success threshold, to generate a resource feature recognition model.

[0045] As a preferred solution, the area determination module is specifically configured to: determine a center point of each resource feature, determine a circumscribed circle of the resource feature with the center point as the center, determine an area value corresponding to the resource feature according to the area of the circumscribed circle; determine a target resource type of the resource feature corresponding to the circumscribed circle according to the resource feature and the corresponding resource type marked by the preset resource feature recognition model; determine a preset weight value according to the target resource type, calculate the product of the area value and the preset weight value as the resource value corresponding to the resource feature.

[0046] As a preferred solution, the image fusion module is specifically configured to: determine the end point of the target remote sensing image of a current segment as a first fusion point in the resource feature marked in each segment of the target remote sensing image respectively; determine the resource boundary corresponding to the end point of the target remote sensing image of the current segment as a second fusion point in the adjacent next segment of the target remote sensing image; align the first fusion point and the second fusion point as a reference resource feature point, and fuse the current segment of the target remote sensing image with the adjacent next segment of the target remote sensing image, until all target remote sensing images are sequentially fused to obtain a long continuous remote sensing image.

[0047] As a preferred solution, the calculation formula of the total value of forest resource assets is:

[0048]

[0049] Wherein, S0 is the total value of forest resource assets; S i is the resource value corresponding to the i th resource feature, k i is the preset type threshold value corresponding to the resource type of the i th resource feature; i is the resource feature, and n is the total number of resource features.

[0050] The application also provides a computer readable storage medium, which comprises a stored computer program; wherein the computer program controls a device where the computer readable storage medium is located to execute the forest resource asset determination method according to any one of the above when running.

[0051] The application also provides a terminal device, which comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, and the processor realizes the forest resource asset determination method according to any one of the above when executing the computer program.

[0052] Compared with the prior art, the application has the following beneficial effects:

[0053] The technical scheme of the application can accurately calculate the total value of forest resource assets in the target forest area range by segmenting the target forest area range for remote sensing image acquisition, sequentially fusing all target remote sensing images after calculating the resource value corresponding to each resource feature, thereby solving the technical problem that the inaccuracy in image acquisition and long continuous image calculation in the prior art causes the strategy of using big data to calculate part of the forest area to only be able to calculate in a small range, and the strategy cannot accurately calculate the forest range with a large area and a wide region; and the strategy can effectively, quickly and accurately determine the actual situation of forest resource assets for a forest range with a large area and a wide region, and provides strong data support for ecological environment protection. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 A step flow chart of a forest resource asset determination method provided by the present application;

[0055] Figure 2 A structural schematic diagram of a forest resource asset determination system provided by the present application;

[0056] Figure 3 A structural schematic diagram of a terminal device provided by the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0058] Embodiment one

[0059] Please refer to Figure 1 A step flow chart of a forest resource asset determination method provided by the present application, comprising steps 101 to 105, each step is specifically as follows:

[0060] Step 101, obtaining target latitude and longitude information of a forest resource asset to be detected, and collecting segmented remote sensing images of a target forest area range within the target latitude and longitude information by a UAV to obtain multiple target remote sensing images.

[0061] In the embodiment, the step 101 specifically comprises: step 1011, determining a maximum collection range according to a model of the UAV, dividing the target latitude and longitude information into multiple latitude and longitude ranges according to the maximum collection range and arranging them in sequence; step 1012, controlling the UAV to collect segmented remote sensing images according to the arranged latitude and longitude ranges in sequence and transmitting the collected target remote sensing images in real time; step 1013, identifying resource boundary features of the target remote sensing images transmitted back in real time, and determining the resource boundary as an end point of the target remote sensing image of the current segment when judging that a distance value between the identified resource boundary and the target remote sensing image of the next segment is less than a preset distance value; and step 1014, re-dividing the latitude and longitude ranges which have not been collected into images according to the maximum collection range and arranging them in sequence until the end point of each target remote sensing image is determined.

[0062] Specifically, in order to solve the problem of inaccuracy in measuring and calculating between image acquisition and long continuous images in the prior art, we use a segmented remote sensing image acquisition method during image acquisition by a UAV. In actual operation, the target latitude and longitude information is used for preliminary segmentation. However, considering that in actual application, the same segment may be divided into two adjacent segments due to the obvious feature boundary of forest resources, which may affect the accuracy of the data, we also need to consider the distance value between the resource boundary of the current segmented image and the target remote sensing image of the next segment during segmentation of the segmented image. That is, when the distance between the forest resource boundary in the current segmented image and the next image is already "very short", the image acquisition for the current segment is stopped at this time, and the current forest resource boundary is taken as the end point of the current segment. In the subsequent forest resource feature collection, the next segmented image is collected until all the forest resource images are collected.

[0063] Step 102, respectively, for each segment of the target remote sensing image feature recognition, mark the resource feature in the target remote sensing image, and get the corresponding resource feature in each segment of the target remote sensing image.

[0064] In the first aspect of the embodiment, the step 102 specifically includes: respectively transmitting each segment of the target remote sensing image as an input image to a preset resource feature recognition model for resource feature recognition, to obtain an output image, wherein the resource feature in the target remote sensing image is marked in the output image; wherein the preset resource feature recognition model is used to identify the resource feature in the input image and determine the resource type, and output the output image marked with the resource feature and the corresponding resource type.

[0065] In the second aspect of the embodiment, the construction process of the preset resource feature recognition model specifically includes: obtaining a historical remote sensing image, marking the forest resource feature in the historical remote sensing image to generate first marking information; determining the corresponding resource type according to the marked forest resource feature to generate second marking information, and associating the first marking information and the second marking information and marking them in the historical remote sensing image to generate a final remote sensing image; constructing an initial recognition model through a neural network algorithm, inputting the final remote sensing image into the initial recognition model for training until the training times reach a times threshold, to obtain a trained recognition model; inputting the historical remote sensing image into the trained recognition model for testing until the test success rate reaches a success threshold to generate a resource feature recognition model.

[0066] Specifically, in order to accurately identify the resource features in the target remote sensing image, including the corresponding resource type, we need to construct and train the model, use the labeled forest resource features in the historical remote sensing image, and determine the corresponding resource type, and then mark the input image of the model, which can accurately train the identification model we need.

[0067] Step 103, respectively calculate the area size of each resource feature in the target remote sensing image where it is located, get the area value, and calculate the resource value corresponding to each resource feature according to the area value.

[0068] In this embodiment, the step 103 specifically includes: step 1031, respectively determine the center point of each resource feature, take the center point as the center of the circumscribed circle, determine the circumscribed circle of the resource feature, and determine the area value corresponding to the resource feature according to the area of the circumscribed circle; Step 1032, according to the output labeled resource feature and its corresponding resource type of the preset resource feature identification model, determine the target resource type of the resource feature corresponding to the circumscribed circle; Step 1033, according to the target resource type, determine the preset weight value, calculate the product of the area value and the preset weight value as the resource value corresponding to the resource feature.

[0069] Specifically, in order to accurately identify the resource value corresponding to each resource feature, we use the calculation method of the circumscribed circle, and combine the product of the preset weight values corresponding to different resource types, which can accurately express the resource value of different types of resource features, so as to facilitate the total value calculation in the subsequent process.

[0070] Step 104, according to the target latitude and longitude information, arrange all the target remote sensing images in order, and respectively determine the reference resource feature point according to the resource feature marked in each section of the target remote sensing image, and sequentially fuse all the target remote sensing images arranged in order according to the reference resource feature point, to get a long continuous remote sensing image.

[0071] In this embodiment, the step 104 specifically includes: step 1041, respectively determine the end point of the current section of the target remote sensing image in the resource feature marked in each section of the target remote sensing image as the first fusion point; Step 1042, determine the resource boundary corresponding to the end point of the current section of the target remote sensing image in the adjacent next section of the target remote sensing image as the second fusion point; Step 1043, align the first fusion point and the second fusion point as the reference resource feature point, and fuse the current section of the target remote sensing image with the adjacent next section of the target remote sensing image, until all the target remote sensing images are sequentially fused to complete, to get a long continuous remote sensing image.

[0072] Specifically, in order to realize the calculation of long continuous images, the most important step after accurately identifying each segmented image is to fuse the segmented images. In actual application, by determining the fusion points, the reference resource feature points are aligned, the current target remote sensing image and the adjacent next target remote sensing image are fused, until all target remote sensing images are sequentially fused to obtain long continuous remote sensing images.

[0073] In step 105, the forest resource asset total value of the target forest region range is calculated according to the resource value corresponding to each resource feature in the long continuous remote sensing image.

[0074] In the embodiment, the calculation formula of the forest resource asset total value is:

[0075]

[0076] Wherein, S0 is the forest resource asset total value; S i is the resource value corresponding to the i th resource feature, k i is the preset type threshold value corresponding to the resource type of the i th resource feature; i is the resource feature, and n is the total number of resource features.

[0077] Specifically, in the actual calculation process, the above formula can objectively and accurately determine the forest resource asset total value of the target forest region range.

[0078] The technical scheme of the present application can accurately calculate the forest resource asset total value of the target forest region range by segmenting the target forest region range for remote sensing image acquisition, sequentially fusing all target remote sensing images arranged in sequence after calculating the resource value corresponding to each resource feature, thereby solving the technical problem that the inaccuracy of image acquisition and long continuous image calculation in the prior art leads to the strategy of using big data to measure part of the forest region only in a small range, and the forest range with large area and wide region cannot be accurately measured. For the forest range with large area and wide region, the actual situation of forest resource assets can be effectively, quickly and accurately determined, which provides strong data support for ecological environment protection.

[0079] Embodiment two

[0080] Please refer to Figure 2 A structural schematic diagram of a forest resource asset determination system provided by another embodiment of the present application, comprising: an image acquisition module, a feature marking module, an area determination module, an image fusion module and a total value calculation module.

[0081] The image acquisition module is configured to acquire target latitude and longitude information of forest resource assets to be detected, and acquire segmented remote sensing images of a target forest area within the target latitude and longitude information by using the UAV.

[0082] In this embodiment, the image acquisition module is specifically configured to determine a maximum acquisition range according to a model of the UAV, divide the target latitude and longitude information into multiple latitude and longitude ranges according to the maximum acquisition range, and arrange the latitude and longitude ranges in sequence; control the UAV to acquire segmented remote sensing images according to the arranged latitude and longitude ranges in sequence, and transmit the acquired target remote sensing images in real time; identify resource boundary features of the target remote sensing images transmitted in real time, and determine the resource boundary as an end point of the target remote sensing image of a current segment when it is determined that a distance value between the identified resource boundary and a next target remote sensing image is less than a preset distance value; redivide latitude and longitude ranges that have not been subjected to image acquisition according to the maximum acquisition range, and arrange the latitude and longitude ranges in sequence until end points of target remote sensing images of all segments are determined.

[0083] The feature marking module is configured to identify features of each of the target remote sensing images, mark resource features in the target remote sensing images, and obtain corresponding resource features in each of the target remote sensing images.

[0084] In the first aspect of this embodiment, the feature marking module is specifically configured to transmit each of the target remote sensing images as an input image to a preset resource feature identification model to identify resource features, and obtain an output image, wherein resource features in the target remote sensing image are marked in the output image; and the preset resource feature identification model is configured to identify resource features in an input image and determine a resource type, and output an output image in which resource features and a corresponding resource type are marked.

[0085] In the second aspect of this embodiment, a construction process of the preset resource feature identification model specifically includes: acquiring historical remote sensing images, marking forest resource features in the historical remote sensing images to generate first marking information; determining corresponding resource types according to the marked forest resource features to generate second marking information, and associating the first marking information and the second marking information to mark in the historical remote sensing images to generate a final remote sensing image; constructing an initial identification model by using a neural network algorithm, inputting the final remote sensing image into the initial identification model to train until a training frequency reaches a frequency threshold to obtain a trained identification model; and inputting the historical remote sensing image into the trained identification model to test until a test success rate reaches a success threshold to generate a resource feature identification model.

[0086] The area determination module is configured to calculate the area size of each resource feature in the target remote sensing image in which the resource feature is located, to obtain an area value, and to calculate a resource value corresponding to each resource feature according to the area value.

[0087] In this embodiment, the area determination module is specifically configured to determine the center point of each resource feature, determine the circumscribed circle of the resource feature with the center point as the center, determine the area value corresponding to the resource feature according to the area of the circumscribed circle, determine the target resource type of the resource feature corresponding to the circumscribed circle according to the resource feature recognition model outputting the resource feature and the corresponding resource type, determine a preset weight value according to the target resource type, and calculate the product of the area value and the preset weight value as the resource value corresponding to the resource feature.

[0088] The image fusion module is configured to arrange all target remote sensing images in order according to the target latitude and longitude information, determine a reference resource feature point according to the resource feature marked in each segment of the target remote sensing image, and sequentially fuse all target remote sensing images arranged in order according to the reference resource feature point, to obtain a long continuous remote sensing image.

[0089] In this embodiment, the image fusion module is specifically configured to determine the end point of the current segment of the target remote sensing image as a first fusion point in the resource feature marked in each segment of the target remote sensing image, determine the resource boundary corresponding to the end point of the current segment of the target remote sensing image as a second fusion point in the adjacent next segment of the target remote sensing image, align the first fusion point and the second fusion point as the reference resource feature point, and fuse the current segment of the target remote sensing image with the adjacent next segment of the target remote sensing image until all target remote sensing images are sequentially fused to obtain a long continuous remote sensing image.

[0090] The total value calculation module is configured to calculate the total value of forest resources assets of the target forest region range according to the resource value corresponding to each resource feature in the long continuous remote sensing image.

[0091] In this embodiment, the calculation formula of the total value of forest resources assets is as follows:

[0092]

[0093] wherein, S0 is the total value of forest resources assets; Si is the resource value corresponding to the i th resource feature, k i is the preset type threshold value corresponding to the resource type of the i th resource feature, i is the resource feature, and n is the total number of resource features. i i

[0094] Embodiment Three​​

[0095] The embodiment of the present application further provides a computer readable storage medium, which comprises a stored computer program; wherein the computer program controls a device where the computer readable storage medium is located to execute the forest resource asset determination method according to any one of the above embodiments when running.

[0096] Embodiment four

[0097] Please refer to Figure 3 is a structural schematic diagram of an embodiment of the terminal device provided by the embodiment of the present application, the terminal device comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, and the processor realizes the forest resource asset determination method according to any one of the above embodiments when executing the computer program.

[0098] Preferably, the computer program can be divided into one or more modules / units (such as computer programs, computer programs), which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the terminal device.

[0099] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can also be any conventional processor. The processor is the control center of the terminal device, and connects various parts of the terminal device through various interfaces and lines.

[0100] The memory mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function, etc., and the data storage area can store relevant data, etc. In addition, the memory can be a high-speed random access memory, and can also be a non-volatile memory such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., or can be other volatile solid-state storage devices.

[0101] It should be noted that the terminal device can include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the terminal device is only an example and does not constitute a limitation on the terminal device, and can include more or fewer components, or combine certain components, or different components.

[0102] The above-described specific embodiments further illustrate the purposes, technical solutions and beneficial effects of the present application. It should be understood that the above-described specific embodiments are only examples of the present application and do not limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A forest resource asset determination method characterized by, The application relates to a method for acquiring forest resource assets, and belongs to the technical field of forest resource assets acquisition. The method comprises the following steps: acquiring target latitude and longitude information of a forest resource asset to be detected, determining a maximum collection range according to a model of a UAV, segmenting the target latitude and longitude information into multiple latitude and longitude ranges according to the maximum collection range, and sequentially arranging the latitude and longitude ranges; controlling the UAV to sequentially collect segmented remote sensing images according to the arranged latitude and longitude ranges, and transmitting the collected target remote sensing images in real time; performing resource boundary feature recognition on the target remote sensing images transmitted in real time, determining a terminal point of a current segment of target remote sensing images when a distance value between a resource boundary and a next segment of target remote sensing images is less than a preset distance value; re-segmenting latitude and longitude ranges that have not been subjected to image collection according to the maximum collection range, sequentially arranging the latitude and longitude ranges, and determining terminal points of target remote sensing images of all segments until the terminal points of the target remote sensing images of all segments are determined; and obtaining multiple target remote sensing images; performing feature recognition on each of the target remote sensing images, marking resource features in the target remote sensing images, and obtaining corresponding resource features in each of the target remote sensing images; calculating an area size of each of the resource features in the target remote sensing image in which the resource feature is located, obtaining an area value, and calculating a resource value corresponding to each of the resource features according to the area value; arranging all the target remote sensing images in sequence according to the target latitude and longitude information, determining a reference resource feature point according to the resource features marked in each of the target remote sensing images, sequentially fusing all the target remote sensing images arranged in sequence according to the reference resource feature point, and obtaining a long continuous remote sensing image; 2. The forest resource asset determination method of claim 1, wherein, calculating a total value of forest resource assets in the target forest region range according to the resource values of the resource features in the long continuous remote sensing image. The step of performing feature recognition on each of the target remote sensing images, marking resource features in the target remote sensing images, and obtaining corresponding resource features in each of the target remote sensing images specifically comprises the following steps: transmitting each of the target remote sensing images as an input image to a preset resource feature recognition model to perform resource feature recognition, and obtaining an output image in which resource features in the target remote sensing image are marked; 3. The forest resource asset determination method of claim 2, wherein, The preset resource feature recognition model is used for recognizing resource features in an input image and determining a resource type, and outputs an output image in which resource features and corresponding resource types are marked. The construction process of the preset resource feature recognition model specifically comprises the following steps: acquiring historical remote sensing images, marking forest resource features in the historical remote sensing images, and generating first marking information; determining corresponding resource types according to the marked forest resource features, generating second marking information, and associating the first marking information and the second marking information to mark the historical remote sensing images, and generating a final remote sensing image; constructing an initial recognition model through a neural network algorithm, inputting the final remote sensing image into the initial recognition model for training until a training frequency reaches a frequency threshold, and obtaining a trained recognition model. The historical remote sensing image is input into the training identification model for testing, and a resource feature identification model is generated when a test success rate reaches a success threshold.

4. The forest resource asset determination method of claim 2, wherein, The step of calculating the area value of each resource feature in the target remote sensing image where the resource feature is located and calculating the resource value corresponding to each resource feature according to the area value specifically includes: A center point of each resource feature is determined, and a circumscribed circle of the resource feature is determined with the center point as the center, and the area value corresponding to the resource feature is determined according to the area of the circumscribed circle; The target resource type of the resource feature corresponding to the circumscribed circle is determined according to the resource feature identification model output labeled with resource features and corresponding resource types; A preset weight value is determined according to the target resource type, and the product of the area value and the preset weight value is calculated as the resource value corresponding to the resource feature.

5. The forest resource asset determination method of claim 1, wherein, The step of determining a reference resource feature point according to the resource features labeled in each section of the target remote sensing image, and sequentially fusing all the target remote sensing images according to the reference resource feature points to obtain a long continuous remote sensing image specifically includes: A terminal point of the current section of the target remote sensing image is determined as a first fusion point in the resource features labeled in each section of the target remote sensing image; A resource boundary corresponding to the terminal point of the current section of the target remote sensing image is determined as a second fusion point in the next section of the target remote sensing image adjacent to the current section; The first fusion point and the second fusion point are aligned as reference resource feature points, and the current section of the target remote sensing image and the next section of the target remote sensing image adjacent to the current section are fused until all the target remote sensing images are sequentially fused to obtain a long continuous remote sensing image.

6. A forest resource asset determination system characterized by, It includes: An image acquisition module, a feature labeling module, an area determination module, an image fusion module, and a total value calculation module; The image acquisition module is configured to obtain target longitude and latitude information of forest resource assets to be detected, determine a maximum acquisition range according to a model of a UAV, divide the target longitude and latitude information into multiple sections of longitude and latitude ranges according to the maximum acquisition range, and arrange the sections sequentially; control the UAV to perform segmented remote sensing image acquisition according to the arranged longitude and latitude ranges, and transmit the acquired target remote sensing images in real time; perform resource boundary feature identification on the target remote sensing images transmitted in real time, and determine a terminal point of a current section of the target remote sensing image when a distance value between the resource boundary and a next section of the target remote sensing image is less than a preset distance value; The remaining longitude and latitude ranges that have not been subjected to image acquisition are re-divided according to the maximum acquisition range, and arranged sequentially until the terminal points of all the sections of the target remote sensing images are determined, and multiple sections of target remote sensing images are obtained; The feature labeling module is configured to perform feature identification on each section of the target remote sensing images respectively, label resource features in the target remote sensing images, and obtain corresponding resource features in each section of the target remote sensing images. The area determining module is configured to calculate the area size of each resource feature in the target remote sensing image in which the resource feature is located to obtain an area value, and calculate a resource value corresponding to each resource feature according to the area value; The image fusion module is configured to arrange all target remote sensing images in order according to the target latitude and longitude information, determine a reference resource feature point according to the resource features marked in each target remote sensing image, and sequentially fuse all target remote sensing images arranged in order according to the reference resource feature point to obtain a long continuous remote sensing image. The total value calculating module is configured to calculate a total value of forest resource assets of the target forest region range according to the resource values corresponding to each resource feature in the long continuous remote sensing image.

7. A computer-readable storage medium, characterized in that, The computer readable storage medium comprises a computer program stored therein; when the computer program is executed, the computer readable storage medium controls a device in which the computer readable storage medium is located to execute the forest resource asset determination method according to any one of claims 1-5.

8. A terminal device, comprising: The device comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the forest resource asset determination method according to any one of claims 1-5 when the computer program is executed.

Images