Forest parameter extraction method, device, equipment and medium based on multi-source remote sensing

By combining hyperspectral images and lidar point cloud data and using multi-source remote sensing technology, the problem of low accuracy of forest structure parameters extraction under traditional single remote sensing technology is solved, and higher accuracy and accuracy are achieved.

CN117765401BActive Publication Date: 2024-12-03AEROSPACE XINDE ZHITU (BEIJING) TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410041945.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-11
Publication Date
2024-12-03
Estimated Expiration
2044-01-11

AI Technical Summary

Technical Problem

The traditional forest structure parameter extraction method uses a single remote sensing technology, which leads to low recognition accuracy and large errors, making it impossible to accurately obtain forest structure parameters.

Method used

Multi-source remote sensing data is used to combine hyperspectral images and lidar point cloud data to determine forest types through hyperspectral images, and basic parameters are extracted using lidar point cloud data to calculate forest structure parameters to improve accuracy.

Benefits of technology

Through the combination of multi-source remote sensing data, the accuracy of forest structural parameters is significantly improved, errors are reduced, and more accurate forest types and structural parameters are provided.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117765401B_ABST
    Figure CN117765401B_ABST
Patent Text Reader

Abstract

The present application relates to the field of forest resource survey, and in particular to a forest parameter extraction method, device, equipment and medium based on multi-source remote sensing. The method includes: obtaining multi-source high-resolution remote sensing data of the target area; the multi-source high-resolution remote sensing data includes hyperspectral image data and laser radar point cloud data; based on the hyperspectral image data, determining the forest type of the target area; based on the laser radar point cloud data, determining the basic parameters of the target area; the basic parameters are the relevant measurement parameters of the trees in the target area; based on the forest type and the basic parameters, determining the forest structure parameters. The present application improves the accuracy of forest structure parameters by combining hyperspectral image technology and laser radar technology.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of forest resource survey, and in particular to a forest parameter extraction method, device, equipment and medium based on multi-source remote sensing. Background Art

[0002] In recent years, with the development of remote sensing technology, the ground resolution of remote sensing satellites has become higher and higher, and the amount of information obtained has become larger and larger. More importantly, the application scope of remote sensing technology is very wide. It can not only be used in military research and urban planning, but also in forest monitoring. Traditional remote sensing technology extracts forest structure parameters mainly using a single remote sensing technology, such as lidar technology. LiDAR is used to identify the type of vegetation, and then obtain the three-dimensional structural characteristics of the trees, from which the position of the trees in the entire forest and the tree height and other information are extracted, and then the forest structure parameters are extracted.

[0003] However, the traditional forest structure parameter extraction method uses a single remote sensing technology and the lidar technology has low accuracy in identifying vegetation types in the forest, resulting in a certain degree of error in the final extracted forest structure parameters. Summary of the invention

[0004] In order to improve the accuracy of forest structure parameters, the present application provides a forest parameter extraction method, device, equipment and medium based on multi-source remote sensing.

[0005] In the first aspect, the present application provides a forest parameter extraction method based on multi-source remote sensing, which adopts the following technical solution:

[0006] Acquire multi-source high-resolution remote sensing data of the target area;

[0007] The multi-source high-resolution remote sensing data includes hyperspectral image data and laser radar point cloud data;

[0008] Determining the forest type of the target area based on the hyperspectral image data;

[0009] Determining basic parameters of the target area based on the laser radar point cloud data;

[0010] The basic parameters are relevant measurement parameters of trees in the target area;

[0011] Based on the forest type and the basic parameters, forest structure parameters are determined.

[0012] By adopting the above technical scheme, multi-source high-resolution remote sensing data of the target area to be processed is obtained, and the multi-source high-resolution remote sensing data not only includes hyperspectral image data, but also includes lidar point cloud data; and due to the wide spectral response range and high spectral resolution of hyperspectral image data, the difference bands between different forest types in the target area are analyzed based on the hyperspectral image data to determine the forest type of the target area; then, by utilizing the high point cloud accuracy, strong laser penetration ability and accurate spatial coordinate information of the lidar, the lidar point cloud data is pre-processed to determine the characteristic points of each vegetation in the forest in the target area, and then the three-dimensional point cloud image of each vegetation is used to determine the basic parameters of the target area; finally, based on the forest type of the target area and the determined basic parameters, the forest structure parameters are determined after further calculation and processing; thereby, the accuracy of the forest structure parameters is improved by combining the hyperspectral image technology and the lidar technology.

[0013] In a possible implementation, the step of acquiring multi-source high-resolution remote sensing data of a target area includes:

[0014] Acquiring initial multi-source high-resolution remote sensing data within the target area;

[0015] Determining whether the initial multi-source high-resolution remote sensing data contains non-forest feature points;

[0016] If the initial multi-source high-resolution remote sensing data contains non-forest feature points, the non-forest feature points are extracted and removed to determine the multi-source high-resolution remote sensing data.

[0017] By adopting the above technical scheme, the initial multi-source high-resolution remote sensing data in the target area is obtained, and the initial multi-source high-resolution remote sensing data is judged to determine whether there are feature points of non-forest vegetation in the initial multi-source high-resolution remote sensing data. If there are feature points of non-forest vegetation in the initial multi-source high-resolution remote sensing data, it indicates that there may be areas without vegetation in the target area, or there may be communities other than vegetation communities, such as water communities. Therefore, the feature points of the non-forest vegetation are extracted, and the feature points of the non-forest vegetation are removed from the initial multi-source high-resolution remote sensing data to determine the multi-source high-resolution remote sensing data; thereby, the initial multi-source high-resolution remote sensing data is screened, the influencing factors are removed, and the processing efficiency is improved.

[0018] In a possible implementation, determining the forest type of the target area based on the hyperspectral image data includes:

[0019] Splitting the hyperspectral image data to determine a plurality of sub-hyperspectral image data;

[0020] Remove noise points from the plurality of sub-hyperspectral image data to determine a plurality of first target sub-hyperspectral image data;

[0021] The plurality of first target sub-hyperspectral image data are divided according to a preset rule to determine the forest type of the target area.

[0022] By adopting the above technical scheme, since the amount of data information is too large, after obtaining multi-source high-resolution remote sensing data, the hyperspectral image data therein is first split into equal proportions to determine multiple sub-hyperspectral image data, and since the spectral imager may cause the grayscale value to be continuously high or low in a certain direction when scanning the object, a sensor failure will cause a noise reduction model to be constructed, and the multiple sub-hyperspectral image data are subjected to noise reduction processing at the same time to determine multiple first target sub-hyperspectral image data, and finally, the multiple first target sub-hyperspectral image data are divided according to preset rules to determine the forest type of the target area; thereby, by correcting the image data, it is convenient to distinguish the forest type.

[0023] In a possible implementation, dividing the plurality of first target sub-hyperspectral image data according to a preset rule to determine the forest type of the target area includes:

[0024] Extracting color information, shape information and texture information respectively contained in the plurality of first target sub-hyperspectral image data;

[0025] Screening the plurality of first target sub-hyperspectral image data step by step according to the color information, the shape information and the texture information to determine a plurality of second target sub-hyperspectral image data;

[0026] The forest type of the target area is determined based on the plurality of second target sub-hyperspectral image data.

[0027] By adopting the above technical scheme, the color information, shape information and texture information respectively contained in the multiple first target sub-hyperspectral image data are extracted, and then the color information, shape information and texture information are used as screening conditions, and the multiple first target sub-hyperspectral image data are screened step by step through the screening conditions to determine the multiple second target sub-hyperspectral image data, that is, the multiple image data with the same color information, the same shape information and the same texture information; finally, based on the multiple second target sub-hyperspectral image data, the forest type of the target area is determined; thereby, the forest type division is made more accurate through multiple screening criteria.

[0028] In a possible implementation manner, the forest type includes a plurality of sub-forest types, and determining the forest type of the target area based on the plurality of second target sub-hyperspectral image data includes:

[0029] Acquire color score information, shape score information, and texture score information respectively corresponding to the plurality of second target sub-hyperspectral image data;

[0030] Substituting the color score information, the shape score information and the texture score information into a preset color score information set, a preset shape score information set and a preset texture score information set for step-by-step matching to determine a plurality of target matching information;

[0031] The target matching information includes color matching information, shape matching information and texture matching information;

[0032] If the same matching object exists in the color matching information, shape matching information, and texture matching information corresponding to any second target sub-hyperspectral image data, the sub-forest type corresponding to the matching object is obtained, and the sub-forest type is used as the sub-forest type of any third target sub-hyperspectral image data;

[0033] A plurality of sub-forest types are aggregated to determine the forest type of the target area.

[0034] By adopting the above technical solution, in order to facilitate matching, the color score information, shape score information and texture score information corresponding to the plurality of third target sub-hyperspectral image data are obtained. In order to determine the forest type specifically corresponding to each second target sub-hyperspectral image data, the plurality of color score information are respectively substituted into the preset color score information set; the plurality of shape score information are respectively substituted into the preset shape score information set for matching; the plurality of texture score information are respectively substituted into the preset texture score information set for matching, and the plurality of target matching information, i.e., the matching information including the color matching information, the shape matching information and the texture matching information, is determined. If the same matching object exists in the color matching information, the shape matching information and the texture matching information corresponding to any second target sub-hyperspectral image data, it indicates that the forest type corresponding to the matching object is the same as the forest type of any second target sub-hyperspectral image data, and therefore the sub-forest type corresponding to the matching object is obtained and used as the sub-forest type of any second target sub-hyperspectral image data; finally, the sub-forest types that are successfully matched are traversed to determine the forest type of the target area; thereby accurately determining the forest type contained in the target area.

[0035] In a possible implementation, determining basic parameters of the target area based on the laser radar point cloud data includes:

[0036] Extract features from the laser radar point cloud data to determine the pixel points corresponding to each tree;

[0037] Based on the pixel points, determining the number of trees in the target area;

[0038] Based on the tree number information and the pixel points, and taking the ground as a base point, determining multiple breast diameter information, multiple tree height information, and multiple crown area information corresponding to multiple trees;

[0039] The multiple breast diameter information, the multiple tree height information and the multiple tree crown area information are summarized to determine the basic parameters.

[0040] By adopting the above technical solution, the lidar point cloud data is feature extracted to determine the pixel points corresponding to each tree, and then the number of trees contained in the target area is determined based on multiple pixel points; since each pixel point has corresponding three-dimensional coordinate information, based on the tree number information and pixel points, and taking the ground as the base point, multiple breast diameter information, multiple tree height information and multiple crown area information corresponding to multiple trees are determined; finally, the breast diameter information, tree height information and crown area information of each tree are summarized to determine the basic parameters of multiple trees; thereby obtaining the basic parameters of the target area, providing a basis for the subsequent extraction of forest structure parameters.

[0041] In a possible implementation manner, determining the forest structure parameter based on the forest type and the basic parameter includes:

[0042] Determine the total forest area information corresponding to the forest type;

[0043] Determine forest density based on the total forest area information and the number of trees;

[0044] Determine forest density based on the total forest area information and the breast diameter information;

[0045] Determining forest canopy density based on the tree crown area information and the tree height information;

[0046] The forest density, the forest sparseness and the forest canopy density are summarized to determine forest structure parameters.

[0047] By adopting the above technical scheme, the total forest area information corresponding to the forest type is determined, and then the forest density corresponding to the forest type is determined through calculation based on the total forest area information and the number of trees corresponding to the forest type; the forest sparseness corresponding to the forest type is determined through calculation based on the total forest area information and the breast diameter information corresponding to the forest type; the forest canopy density of the forest type is determined through calculation based on the crown area information and the tree height information corresponding to the forest type; finally, the forest density, forest sparseness and forest canopy density corresponding to the forest type are summarized to determine the forest structure parameters; thus, the final forest structure parameters are obtained through calculation through the basic parameters.

[0048] In the second aspect, the present application provides a forest parameter extraction device based on multi-source remote sensing, which adopts the following technical solution:

[0049] A forest parameter extraction device based on multi-source remote sensing includes: a remote sensing data acquisition module, a forest type determination module, a basic parameter determination module and a structural parameter determination module, wherein:

[0050] Remote sensing data acquisition module, used to obtain multi-source high-resolution remote sensing data of the target area;

[0051] The multi-source high-resolution remote sensing data includes hyperspectral image data and laser radar point cloud data;

[0052] A forest type determination module, used to determine the forest type of the target area based on the hyperspectral image data;

[0053] A basic parameter determination module, used to determine the basic parameters of the target area based on the laser radar point cloud data;

[0054] The basic parameters are relevant measurement parameters of trees in the target area;

[0055] The structural parameter determination module is used to determine the forest structural parameters based on the forest type and the basic parameters.

[0056] By adopting the above technical scheme, the remote sensing data acquisition module obtains multi-source high-resolution remote sensing data of the target area to be processed, and the multi-source high-resolution remote sensing data not only includes hyperspectral image data, but also includes laser radar point cloud data; due to the wide spectral response range and high spectral resolution of hyperspectral image data, the forest type determination module then analyzes the difference bands between different forest types in the target area based on the hyperspectral image data to determine the forest type of the target area; then the basic parameter determination module uses the high point cloud accuracy, strong laser penetration ability and accurate spatial coordinate information of the laser radar to pre-process the laser radar point cloud data, and then determine the characteristic points of each vegetation in the forest in the target area, and then use the three-dimensional point cloud image data of each vegetation to determine the basic parameters of the target area; finally, the structural parameter determination module determines the forest structural parameters based on the forest type of the target area and the determined basic parameters after further calculation and processing; thereby, the accuracy of forest structural parameters is improved by combining hyperspectral image technology and laser radar technology.

[0057] In a possible implementation, the remote sensing data acquisition module includes: an initial data acquisition unit, a judgment unit, and a data determination unit, wherein:

[0058] An initial data acquisition unit, used to acquire initial multi-source high-resolution remote sensing data in the target area;

[0059] A judging unit, used for judging whether the initial multi-source high-resolution remote sensing data contains non-forest feature points;

[0060] The data determination unit is used to extract the non-forest feature points if the initial multi-source high-resolution remote sensing data contains the non-forest feature points, and remove the non-forest feature points to determine the multi-source high-resolution remote sensing data.

[0061] In a possible implementation, the forest type determination module includes: a sub-image data determination unit, a target sub-image data determination unit and a forest type unit, wherein:

[0062] A sub-image data determining unit, used for splitting the hyperspectral image data to determine a plurality of sub-hyperspectral image data;

[0063] a target sub-image data determining unit, configured to remove noise points from the plurality of sub-hyperspectral image data to determine a plurality of first target sub-hyperspectral image data;

[0064] The forest type unit is used to divide the plurality of first target sub-hyperspectral image data according to a preset rule to determine the forest type of the target area.

[0065] In a possible implementation, the forest type unit is specifically used to:

[0066] Extracting color information, shape information and texture information respectively contained in the plurality of first target sub-hyperspectral image data;

[0067] Screening the plurality of first target sub-hyperspectral image data step by step according to the color information, the shape information and the texture information to determine a plurality of second target sub-hyperspectral image data;

[0068] The forest type of the target area is determined based on the plurality of second target sub-hyperspectral image data.

[0069] In a possible implementation manner, determining the forest type of the target area based on the plurality of second target sub-hyperspectral image data is specifically used for:

[0070] Acquire color score information, shape score information, and texture score information respectively corresponding to the plurality of second target sub-hyperspectral image data;

[0071] Substituting the color score information, the shape score information and the texture score information into a preset color score information set, a preset shape score information set and a preset texture score information set for step-by-step matching to determine a plurality of target matching information;

[0072] The target matching information includes color matching information, shape matching information and texture matching information;

[0073] If the same matching object exists in the color matching information, shape matching information, and texture matching information corresponding to any second target sub-hyperspectral image data, the sub-forest type corresponding to the matching object is obtained, and the sub-forest type is used as the sub-forest type of any second target sub-hyperspectral image data;

[0074] A plurality of sub-forest types are aggregated to determine the forest type of the target area.

[0075] In a possible implementation, the basic parameter determination module includes: a pixel point determination unit, a tree number information determination unit, a basic information determination unit and a basic parameter determination unit, wherein:

[0076] A pixel point determination unit, used to extract features from the laser radar point cloud data and determine the pixel points corresponding to each tree;

[0077] A tree number information determination unit, used to determine the tree number information of the target area based on the pixel points;

[0078] A basic information determination unit, configured to determine a plurality of breast diameter information, a plurality of tree height information, and a plurality of tree crown area information corresponding to a plurality of trees based on the tree number information and the pixel points and taking the ground as a base point;

[0079] The basic parameter determination unit is used to summarize the multiple breast diameter information, the multiple tree height information and the multiple tree crown area information to determine the basic parameters.

[0080] In a possible implementation, the forest parameter extraction device based on multi-source remote sensing further includes: a sub-area information determination module, a forest density determination module, a forest sparseness determination module, a forest canopy density determination module and a forest structure parameter determination module, wherein:

[0081] A sub-area information determination module is used to determine the total forest area information corresponding to the forest type;

[0082] A forest density determination module, used to determine the forest density based on the total forest area information and the number of trees;

[0083] A forest density determination module, used to determine the forest density based on the total forest area information and the breast diameter information;

[0084] A forest canopy closure determination module, used to determine the forest canopy closure based on the tree crown area information and the tree height information;

[0085] The forest structure parameter determination module is used to summarize the forest density, the forest sparseness and the forest canopy density to determine the forest structure parameters.

[0086] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0087] An electronic device, comprising:

[0088] at least one processor;

[0089] Memory;

[0090] At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned forest parameter extraction method based on multi-source remote sensing.

[0091] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0092] A computer-readable storage medium includes: a computer program that can be loaded by a processor and execute the above-mentioned forest parameter extraction method based on multi-source remote sensing.

[0093] In summary, this application includes the following beneficial technical effects:

[0094] Firstly, the multi-source high-resolution remote sensing data of the target area to be processed is obtained, and the multi-source high-resolution remote sensing data includes not only hyperspectral image data but also lidar point cloud data; due to the wide spectral response range and high spectral resolution of hyperspectral image data, the difference bands between different forest types in the target area are analyzed based on the hyperspectral image data to determine the forest type of the target area; then, the lidar point cloud data is pre-processed by utilizing the high point cloud accuracy, strong laser penetration ability and accurate spatial coordinate information of the lidar to determine the characteristic points of each vegetation in the forest in the target area, and then the three-dimensional point cloud image of each vegetation is used to determine the basic parameters of the target area; finally, based on the forest type of the target area and the determined basic parameters, the forest structure parameters are determined after further calculation and processing; thereby, the accuracy of the forest structure parameters is improved by combining the hyperspectral image technology and the lidar technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 It is a flowchart of a forest parameter extraction method based on multi-source remote sensing according to an embodiment of the present application;

[0096] Figure 2 is a block diagram of a forest parameter extraction device based on multi-source remote sensing according to an embodiment of the present application;

[0097] Figure 3 It is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0098] The following is combined with Figure 1-3 This application is described in further detail.

[0099] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0100] The embodiment of the present application provides a forest parameter extraction method based on multi-source remote sensing, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, a server cluster or a distributed device composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application.

[0101] Reference Figure 1 The method comprises: step S101, step S102, step S103 and step S104, wherein:

[0102] S101. Acquire multi-source high-resolution remote sensing data of a target area, where the multi-source high-resolution remote sensing data includes hyperspectral image data and lidar point cloud data.

[0103] Specifically, the electronic equipment determines the target area where forest structure parameters need to be extracted by determining the four intersection points of the forest area. Then, by measuring the weather conditions, the drone's collection frequency, flight speed and flight altitude are set. The drone traverses the entire target area and scans to obtain multi-source high-resolution remote sensing data of the target area.

[0104] It should be noted that when collecting multi-source high-resolution remote sensing data, several sets of multi-source high-resolution remote sensing data can be collected. On the one hand, relatively clear remote sensing data can be obtained through screening. On the other hand, by comparing and analyzing several sets of multi-source high-resolution remote sensing data, flying objects in the target area can be identified to eliminate the interference of flying objects on the image data.

[0105] S102: Determine the forest type of the target area based on the hyperspectral image data.

[0106] Specifically, the electronic device extracts features from the hyperspectral image data based on the hyperspectral image data in the multi-source high-resolution remote sensing data, and obtains feature information of different forest types in the target area through layer-by-layer division. The electronic device then determines multiple sub-forest types corresponding to the target area based on the feature information. Finally, the electronic device aggregates the multiple sub-forest types to determine the forest type of the target area. In this way, the recognition characteristics of the hyperspectral image data are utilized to more accurately distinguish forest types.

[0107] S103: Determine basic parameters of the target area based on the laser radar point cloud data.

[0108] In the embodiment of the present application, the basic parameters are related measurement parameters of trees in the target area.

[0109] Specifically, since the lidar point cloud data can clearly show the three-dimensional coordinate information of vegetation, the electronic device determines the basic parameters of each tree based on the lidar point cloud data in the multi-source high-resolution remote sensing data, and further the basic parameters of trees of different forest types in the target area; thereby utilizing the characteristics of the lidar to determine the basic parameters of the trees.

[0110] S104. Determine forest structure parameters based on forest type and basic parameters.

[0111] Specifically, the electronic device determines the relevant measurement parameters of the trees corresponding to each sub-forest type based on the basic parameters of the trees determined by the lidar and the forest type determined by the hyperspectral image, and then obtains the sub-forest structure parameters of the area corresponding to each sub-forest type. The sub-forest structure parameters corresponding to all sub-forest types in the target area are then summarized to determine the forest structure parameters of the entire target area; thereby improving the efficiency of forest structure parameter extraction by dividing the area and performing parallel processing.

[0112] The embodiment of the present application provides a forest parameter extraction method based on multi-source remote sensing. The electronic device determines the target area where forest structure parameters need to be extracted by determining four intersection points of the forest area range. Then, by measuring the weather conditions, the drone's acquisition frequency, flight speed and flight altitude are set. The drone traverses the entire target area and scans to obtain multi-source high-resolution remote sensing data of the target area. The electronic device extracts features from the hyperspectral image data based on the hyperspectral image data in the multi-source high-resolution remote sensing data, and obtains feature information of different forest types in the target area through layer-by-layer division. Then, the electronic device determines multiple sub-forest types corresponding to the target area based on the feature information. Finally, the electronic device summarizes the multiple sub-forest types to determine the forest type of the target area. type; since the laser radar point cloud data can clearly see the three-dimensional coordinate information of the vegetation, the electronic device determines the basic parameters of each tree based on the laser radar point cloud data in the multi-source high-resolution remote sensing data, and further the basic parameters of the trees of different forest types in the target area; the electronic device determines the relevant measurement parameters of the trees corresponding to each sub-forest type based on the basic parameters of the trees determined by the laser radar and the forest type determined by the hyperspectral image, and then obtains the sub-forest structure parameters of the area corresponding to each sub-forest type, and then summarizes the sub-forest structure parameters corresponding to all sub-forest types in the target area to determine the forest structure parameters of the entire target area; thereby, the accuracy of the forest structure parameters is improved through the combination of hyperspectral image technology and laser radar technology.

[0113] In step S101, multi-source high-resolution remote sensing data of the target area is obtained, specifically including: obtaining initial multi-source high-resolution remote sensing data in the target area; determining whether the initial multi-source high-resolution remote sensing data contains non-forest feature points; if the initial multi-source high-resolution remote sensing data contains non-forest feature points, extracting the non-forest feature points and removing the non-forest feature points to determine the multi-source high-resolution remote sensing data.

[0114] In the embodiment of the present application, non-forest feature points represent other community feature points except vegetation community feature points.

[0115] Specifically, after determining the target area, the electronic device obtains the initial multi-source high-resolution remote sensing data in the target area, and then the electronic device identifies multiple feature points in the target area. If a part of the target area only contains feature points in two-dimensional space, indicating that no vegetation is planted in the part of the area, the feature points in the part of the area are determined as non-forest feature points. If the initial multi-source high-resolution remote sensing data does not contain non-forest feature points, it indicates that the target area is composed of vegetation communities and there are no interference factors. Therefore, the electronic device directly uses the initial multi-source high-resolution remote sensing data as multi-source high-resolution remote sensing data; if the initial multi-source high-resolution remote sensing data contains non-forest feature points, it indicates that there are not only vegetation communities in the target area, but also some other communities such as water communities, or there is a small part of the target area without vegetation, and the image data of the other communities and the small part of the area are invalid data, so the electronic device removes the invalid data and only retains the feature points of the vegetation community, and then determines the multi-source high-resolution remote sensing data; thereby screening the initial multi-source high-resolution remote sensing data, removing the influencing factors, and improving the processing efficiency.

[0116] Furthermore, based on the hyperspectral image data, the forest type of the target area is determined, specifically including: splitting the hyperspectral image data to determine a plurality of sub-hyperspectral image data; removing noise points from the plurality of sub-hyperspectral image data to determine a plurality of first target sub-hyperspectral image data; dividing the plurality of first target sub-hyperspectral image data according to preset rules to determine the forest type of the target area.

[0117] Specifically, since the forest target area is large and the corresponding multi-source high-resolution remote sensing data of the forest target area is also huge, the electronic device divides the hyperspectral image data contained in the multi-source high-resolution remote sensing data into equal proportions to determine a plurality of sub-hyperspectral image data; since the spectral imager may cause the grayscale value to be continuously high or low in a certain direction when scanning the object, the electronic device constructs a noise reduction model and performs noise reduction processing on the plurality of sub-hyperspectral image data at the same time to determine a plurality of corrected first target sub-hyperspectral image data; finally, the electronic device divides the plurality of first target sub-hyperspectral image data according to preset rules to determine the forest type of the target area; thereby, by correcting the image data, it is convenient to distinguish the forest type.

[0118] Furthermore, the multiple first target sub-hyperspectral image data are divided according to preset rules to determine the forest type of the target area, specifically including: extracting color information, shape information and texture information respectively contained in the multiple first target sub-hyperspectral image data; screening the multiple first target sub-hyperspectral image data step by step according to the color information, the shape information and the texture information to determine multiple second target sub-hyperspectral image data; and determining the forest type of the target area based on the multiple second target sub-hyperspectral image data.

[0119] Specifically, the electronic device extracts the color information, shape information and texture information respectively contained in the multiple target sub-hyperspectral image data, and then uses the color information, shape information and texture information as screening conditions to screen the multiple first target sub-hyperspectral image data. First, the electronic device uses the color information as the first screening condition to screen out the image data with the same color information in groups. Secondly, the electronic device screens each group of image data with the same color information again with the shape information as the second screening condition to determine multiple groups of image data with the same color information and shape information. Finally, the electronic device screens the multiple groups of image data with the same color information and shape information for the last time with the texture information as the third screening condition, and finally determines multiple second target sub-hyperspectral image data, that is, multiple groups of sub-hyperspectral image data with the same color information, multiple groups of shape information and multiple groups of texture information. Then, the electronic device determines the forest type of the target area according to the second target sub-hyperspectral image data. Thus, through multiple screening conditions, the forest type division is made more accurate.

[0120] It should be noted that, since the target sub-hyperspectral image data are obtained by dividing the electronic device in equal proportion, if under the screening condition of color information, any first target sub-hyperspectral image data has different color information, then any first target sub-hyperspectral image data will be further screened so that each first target sub-hyperspectral image data is image data with only one color information.

[0121] Furthermore, based on multiple second target sub-hyperspectral image data, determining the forest type of the target area specifically includes: obtaining color score information, shape score information, and texture score information respectively corresponding to the multiple second target sub-hyperspectral image data; substituting the color score information, the shape score information, and the texture score information into a preset color score information set, a preset shape score information set, and a preset texture score information set for step-by-step matching to determine multiple target matching information; if the same matching object exists in the color matching information, the shape matching information, and the texture matching information corresponding to any second target sub-hyperspectral image data, obtaining the sub-forest type corresponding to the matching object, and using the sub-forest type as the sub-forest type of any second target sub-hyperspectral image data; and summarizing multiple sub-forest types to determine the forest type of the target area.

[0122] In the embodiment of the present application, the forest type includes multiple sub-forest types; the target matching information includes color matching information, shape matching information and texture matching information.

[0123] Specifically, the electronic device assigns color information, shape information, and texture information of the acquired multiple second target sub-hyperspectral image data, determines color score information, shape score information, and texture score information respectively corresponding to the multiple second target sub-hyperspectral image data; substitutes the multiple color score information into a preset color score information set for matching, and if any color score information is equal to any preset color score information in the preset color score information set, the electronic device determines the color matching information.

[0124] Then, the electronic device respectively substitutes the multiple shape score information into the preset shape score information set for matching. If any shape score information is equal to any preset shape score information in the preset shape score information set, the electronic device determines the shape matching information. Finally, the multiple texture score information is respectively substituted into the preset texture score information set for matching. If any texture score information is equal to any preset texture score information in the preset texture score information set, the electronic device determines the texture matching information.

[0125] The electronic device traverses all color matching information, shape matching information and texture matching information. If the same matching object exists in the color matching information, shape matching information and texture matching information corresponding to any second target sub-hyperspectral image data, it indicates that the forest type corresponding to the matching object is the same as the forest type of any second target sub-hyperspectral image data. The electronic device obtains the forest type corresponding to the matching object and uses it as the sub-forest type corresponding to any second target sub-hyperspectral image data. Finally, the electronic device summarizes all successfully matched sub-forest types to determine the forest type of the target area, thereby accurately determining the forest type contained in the target area.

[0126] In step S103, basic parameters of the target area are determined based on the laser radar point cloud data, specifically including: extracting features from the laser radar point cloud data to determine the pixel points corresponding to each tree; determining the number of trees in the target area based on the pixel points; determining multiple breast diameter information, multiple tree height information, and multiple crown area information corresponding to multiple trees based on the tree number information and pixel points, and taking the ground as a base point; and summarizing multiple breast diameter information, multiple tree height information, and multiple crown area information to determine basic parameters.

[0127] Specifically, the electronic device extracts features from the lidar point cloud data, determines the pixel points corresponding to each tree, and determines the number of trees in the target area based on the pixel points of each tree on the ground; then, based on the tree number information and the pixel points of the trees, and taking the ground as the base point, the three-dimensional coordinate information of the pixel points is used to determine the tree height information through the coordinate information of the highest pixel point of each tree, and the breast diameter information is determined through the pixel points of each tree at breast height, and the crown area information is determined by analyzing the pixel points of each tree's crown; finally, the electronic device summarizes all the breast diameter information, tree height information, and crown area information to determine the basic parameters of all trees; thereby obtaining the basic parameters of the target area, providing a basis for the subsequent extraction of forest structure parameters.

[0128] Furthermore, based on the forest type and basic parameters, the forest structure parameters are determined, including: determining the total forest area information corresponding to the forest type; determining the forest density based on the total forest area information and the number of trees; determining the forest sparseness based on the total forest area information and the breast diameter information; determining the forest canopy density based on the crown area information and the tree height information; and summarizing the forest density, forest sparseness and forest canopy density to determine the forest structure parameters.

[0129] Specifically, after the electronic device determines the total forest area information corresponding to the forest type in the target area, it determines the forest sub-area information corresponding to each sub-forest type, and then the electronic device calculates the ratio of the number of trees corresponding to each sub-forest type to the forest sub-area information to determine the forest density corresponding to the sub-forest type; the electronic device uses the breast diameter information to determine the breast height sectional area of ​​each tree corresponding to the sub-forest type, and then sums the breast height sectional areas of multiple trees to determine the total breast height sectional area of ​​the sub-forest type, and then obtains the sum of the breast height sectional areas of the standard forest stands with the highest productivity of the same tree species, and calculates the total breast height sectional area. The ratio is calculated with the sum of the cross-sectional areas at breast height of the standard forest stands with the highest productivity of the same tree species to determine the forest density corresponding to the sub-forest type; finally, the electronic device determines the vertical projection area of ​​the crown through the crown area information of the sub-forest type, and then calculates the ratio of the vertical projection area of ​​the crown to the corresponding forest sub-area information to determine the forest density corresponding to the sub-forest type; finally, the electronic device summarizes the forest density, forest density and forest density corresponding to each sub-forest type to determine the forest structure parameters of the target area; thus, the final forest structure parameters are obtained through calculation and processing of the basic parameters.

[0130] The forest parameter extraction device 20 based on multi-source remote sensing may specifically include: a remote sensing data acquisition module 201, a forest type determination module 202, a basic parameter determination module 203 and a structural parameter determination module 204, wherein:

[0131] A remote sensing data acquisition module 201 is used to acquire multi-source high-resolution remote sensing data of a target area;

[0132] Multi-source high-resolution remote sensing data includes hyperspectral image data and lidar point cloud data;

[0133] A forest type determination module 202 is used to determine the forest type of the target area based on the hyperspectral image data;

[0134] A basic parameter determination module 203 is used to determine basic parameters of the target area based on the laser radar point cloud data;

[0135] The basic parameters are the relevant measurement parameters of the trees in the target area;

[0136] The structural parameter determination module 204 is used to determine the forest structural parameters based on the forest type and basic parameters.

[0137] In a possible implementation of the embodiment of the present application, the remote sensing data acquisition module 201 includes: an initial data acquisition unit, a judgment unit and a data determination unit, wherein:

[0138] An initial data acquisition unit, used to acquire initial multi-source high-resolution remote sensing data in a target area;

[0139] A judgment unit, used to judge whether the initial multi-source high-resolution remote sensing data contains non-forest feature points;

[0140] The data determination unit is used to extract the non-forest feature points if the initial multi-source high-resolution remote sensing data contains non-forest feature points, and remove the non-forest feature points to determine the multi-source high-resolution remote sensing data.

[0141] In a possible implementation of the embodiment of the present application, the forest type determination module 202 includes: a sub-image data determination unit, a target sub-image data determination unit and a forest type unit, wherein:

[0142] A sub-image data determination unit, used for splitting the hyperspectral image data to determine a plurality of sub-hyperspectral image data;

[0143] A target sub-image data determining unit, used to remove noise points from the plurality of sub-hyperspectral image data to determine a plurality of first target sub-hyperspectral image data;

[0144] The forest type unit is used to divide the plurality of first target sub-hyperspectral image data according to preset rules to determine the forest type of the target area.

[0145] In a possible implementation manner of the embodiment of the present application, the forest type unit is specifically used to:

[0146] Extracting color information, shape information and texture information respectively contained in the plurality of first target sub-hyperspectral image data;

[0147] Screening the plurality of first target sub-hyperspectral image data step by step according to the color information, the shape information and the texture information to determine a plurality of second target sub-hyperspectral image data;

[0148] The forest type of the target area is determined based on the plurality of second target sub-hyperspectral image data.

[0149] A possible implementation of the embodiment of the present application is to determine the forest type of the target area based on a plurality of second target sub-hyperspectral image data, specifically for:

[0150] Acquire color score information, shape score information, and texture score information respectively corresponding to the plurality of second target sub-hyperspectral image data;

[0151] Substituting the color score information, the shape score information and the texture score information into a preset color score information set, a preset shape score information set and a preset texture score information set for step-by-step matching to determine a plurality of target matching information;

[0152] The target matching information includes color matching information, shape matching information and texture matching information;

[0153] If the same matching object exists in the color matching information, shape matching information, and texture matching information corresponding to any second target sub-hyperspectral image data, the sub-forest type corresponding to the matching object is obtained, and the sub-forest type is used as the sub-forest type of any second target sub-hyperspectral image data;

[0154] A plurality of sub-forest types are aggregated to determine the forest type of the target area.

[0155] In a possible implementation of the embodiment of the present application, the basic parameter determination module 203 includes: a pixel point determination unit, a tree number information determination unit, a basic information determination unit and a basic parameter determination unit, wherein:

[0156] A pixel point determination unit is used to extract features from the laser radar point cloud data and determine the pixel points corresponding to each tree;

[0157] A tree number information determination unit, used to determine the tree number information of the target area based on the pixel points;

[0158] A basic information determination unit, for determining multiple breast diameter information, multiple tree height information, and multiple crown area information corresponding to multiple trees based on the tree number information and pixel points and taking the ground as a base point;

[0159] The basic parameter determination unit is used to aggregate multiple breast diameter information, multiple tree height information and multiple crown area information to determine basic parameters.

[0160] A possible implementation of the embodiment of the present application is a forest parameter extraction device 20 based on multi-source remote sensing, further comprising: a sub-area information determination module, a forest density determination module, a forest sparseness determination module, a forest canopy density determination module, and a forest structure parameter determination module, wherein:

[0161] A sub-area information determination module is used to determine the total forest area information corresponding to the forest type;

[0162] A forest density determination module, used to determine the forest density based on the total forest area information and the number of trees;

[0163] A forest density determination module is used to determine the forest density based on the total forest area information and the breast diameter information;

[0164] A forest canopy closure determination module is used to determine the forest canopy closure based on the canopy area information and the tree height information;

[0165] The forest structure parameter determination module is used to summarize forest density, forest sparseness and forest canopy density to determine forest structure parameters.

[0166] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0167] The present application also introduces an electronic device from the perspective of a physical device, such as Figure 3 As shown, Figure 3 The electronic device 30 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the electronic device 30 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the electronic device 30 does not constitute a limitation on the embodiments of the present application.

[0168] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0169] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0170] The memory 303 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0171] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment.

[0172] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0173] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0174] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A forest parameter extraction method based on multi-source remote sensing, characterized in that: include: Acquire multi-source high-resolution remote sensing data of the target area; The multi-source high-resolution remote sensing data includes hyperspectral image data and laser radar point cloud data; Determining the forest type of the target area based on the hyperspectral image data; Determining basic parameters of the target area based on the laser radar point cloud data; The basic parameters are relevant measurement parameters of trees in the target area; Determining forest structure parameters based on the forest type and the basic parameters; Determining the forest type of the target area based on the hyperspectral image data includes: Splitting the hyperspectral image data to determine a plurality of sub-hyperspectral image data; Remove noise points from the plurality of sub-hyperspectral image data to determine a plurality of first target sub-hyperspectral image data; Dividing the plurality of first target sub-hyperspectral image data according to a preset rule to determine the forest type of the target area; The step of screening the plurality of first target sub-hyperspectral image data according to a preset rule to determine the forest type of the target area includes: Extracting color information, shape information and texture information respectively contained in the plurality of first target sub-hyperspectral image data; Screening the plurality of first target sub-hyperspectral image data step by step according to the color information, the shape information and the texture information to determine a plurality of second target sub-hyperspectral image data; Determining the forest type of the target area based on the plurality of second target sub-hyperspectral image data; The forest type includes a plurality of sub-forest types, and determining the forest type of the target area based on the plurality of second target sub-hyperspectral image data includes: Obtaining color score information, shape score information, and texture score information respectively corresponding to the plurality of second target sub-hyperspectral image data; Substituting the color score information, the shape score information and the texture score information into a preset color score information set, a preset shape score information set and a preset texture score information set for step-by-step matching to determine a plurality of target matching information; The target matching information includes color matching information, shape matching information and texture matching information; If the same matching object exists in the color matching information, shape matching information, and texture matching information corresponding to any second target sub-hyperspectral image data, the sub-forest type corresponding to the matching object is obtained, and the sub-forest type is used as the sub-forest type of any second target sub-hyperspectral image data; A plurality of sub-forest types are aggregated to determine the forest type of the target area.

2. The forest parameter extraction method based on multi-source remote sensing according to claim 1 is characterized in that: The method of obtaining multi-source high-resolution remote sensing data of the target area includes: Acquiring initial multi-source high-resolution remote sensing data within the target area; Determining whether the initial multi-source high-resolution remote sensing data contains non-forest feature points; If the initial multi-source high-resolution remote sensing data contains non-forest feature points, the non-forest feature points are extracted and removed to determine the multi-source high-resolution remote sensing data.

3. The forest parameter extraction method based on multi-source remote sensing according to claim 1 is characterized in that: Determining basic parameters of the target area based on the laser radar point cloud data includes: Extract features from the laser radar point cloud data to determine the pixel points corresponding to each tree; Based on the pixel points, determining the number of trees in the target area; Based on the tree number information and the pixel points, and taking the ground as a base point, determining multiple breast diameter information, multiple tree height information, and multiple crown area information corresponding to multiple trees; The multiple breast diameter information, the multiple tree height information and the multiple tree crown area information are summarized to determine the basic parameters.

4. The method for extracting forest parameters based on multi-source remote sensing according to claim 3 is characterized in that: Based on the forest type and the basic parameters, forest structure parameters are determined, including: Determine the total forest area information corresponding to the forest type; Determine forest density based on the total forest area information and the number of trees; Determine forest density based on the total forest area information and the breast diameter information; Determining forest canopy density based on the tree crown area information and the tree height information; The forest density, the forest sparseness and the forest canopy density are summarized to determine forest structure parameters.

5. A forest parameter extraction device based on multi-source remote sensing, characterized in that: include: Remote sensing data acquisition module, used to obtain multi-source high-resolution remote sensing data of the target area; The multi-source high-resolution remote sensing data includes hyperspectral image data and laser radar point cloud data; A forest type determination module, used to determine the forest type of the target area based on the hyperspectral image data; A basic parameter determination module, used to determine the basic parameters of the target area based on the laser radar point cloud data; The basic parameters are relevant measurement parameters of trees in the target area; A structural parameter determination module, used to determine forest structural parameters based on the forest type and the basic parameters; The forest type determination module includes: a sub-image data determination unit, a target sub-image data determination unit and a forest type unit, wherein: A sub-image data determining unit, used for splitting the hyperspectral image data to determine a plurality of sub-hyperspectral image data; a target sub-image data determining unit, configured to remove noise points from the plurality of sub-hyperspectral image data to determine a plurality of first target sub-hyperspectral image data; A forest type unit, used for dividing the plurality of first target sub-hyperspectral image data according to a preset rule to determine the forest type of the target area; The forest type unit is specifically used for: Extracting color information, shape information and texture information respectively contained in the plurality of first target sub-hyperspectral image data; Screening the plurality of first target sub-hyperspectral image data step by step according to the color information, the shape information and the texture information to determine a plurality of second target sub-hyperspectral image data; Determining the forest type of the target area based on the plurality of second target sub-hyperspectral image data; The determining the forest type of the target area based on the plurality of second target sub-hyperspectral image data is specifically used for: Obtaining color score information, shape score information, and texture score information respectively corresponding to the plurality of second target sub-hyperspectral image data; Substituting the color score information, the shape score information and the texture score information into a preset color score information set, a preset shape score information set and a preset texture score information set for step-by-step matching to determine a plurality of target matching information; The target matching information includes color matching information, shape matching information and texture matching information; If the same matching object exists in the color matching information, shape matching information, and texture matching information corresponding to any second target sub-hyperspectral image data, the sub-forest type corresponding to the matching object is obtained, and the sub-forest type is used as the sub-forest type of any second target sub-hyperspectral image data; A plurality of sub-forest types are aggregated to determine the forest type of the target area.

6. An electronic device, characterized in that: The electronic device includes: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute a forest parameter extraction method based on multi-source remote sensing as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed in a computer, the computer is caused to execute the forest parameter extraction method based on multi-source remote sensing as described in any one of claims 1 to 4.