Forest terrain remote sensing surveying and mapping method and system for smart forestry

Abstract

The invention provides a forest terrain remote sensing surveying and mapping method and system for smart forestry, and relates to the field of data processing. The method comprises the following steps: acquiring a remote sensing image of a forest region to be measured; the elevation value of each monitoring station in the forest region to be measured and the gray value of the monitoring station at the corresponding position of the remote sensing image are acquired, and an elevation-gray relation model is established; for each partition of the forest region to be measured, determining the vegetation coverage density of the partition based on the gray value of each pixel point in the remote sensing image corresponding to the partition; determining a theoretical elevation value corresponding to the gray value of each pixel point in the remote sensing image corresponding to the partition based on an elevation-gray relation model, and correcting the measured elevation value based on the vegetation coverage density and the theoretical elevation value to obtain a weighted elevation value; and determining a seasonal weight based on the air temperature of the forest region to be measured, and determining a target elevation value of the partition based on the seasonal weight and the weighted elevation value. According to the invention, the accuracy of the elevation data of the forest region can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a forest terrain remote sensing mapping method and system for intelligent forestry. Background Art

[0002] With the development of information technology, especially the maturity of technologies such as remote sensing, Geographic Information System (GIS), and Global Positioning System, the digital and intelligent management of forest resources has also developed in a more refined direction. For example, intelligent forestry can manage, monitor, protect, and develop forest resources based on technologies such as big data, the Internet of Things, artificial intelligence, and remote sensing. By using modern remote sensing technology, data processing technology, and intelligent decision-making systems, it provides strong support for the precise monitoring, management, and protection of forest resources.

[0003] The above digital and intelligent management of forest resources involves remote sensing mapping of forest terrain. In this process, related technologies usually use Light Detection and Ranging (LiDAR) technology to obtain elevation data of forest areas, and then conduct terrain mapping based on the elevation data. However, since the forest surface is generally covered by trees, especially in areas with dense forest canopies and thick tree crowns, it is difficult for optical remote sensing equipment to penetrate the tree crowns to reach the ground, resulting in missing or inaccurate detection results of surface elevation information for the corresponding areas, affecting the accuracy of forest terrain mapping. Summary of the Invention

[0004] In order to solve the problem in related technologies that since the forest surface is generally covered by trees, especially in areas with dense forest canopies and thick tree crowns, it is difficult for optical remote sensing equipment to penetrate the tree crowns to reach the ground, resulting in missing or inaccurate detection results of surface elevation information for the corresponding areas, affecting the accuracy of forest terrain mapping, the present invention provides a forest terrain remote sensing mapping method for intelligent forestry. The specific technical solutions adopted are as follows: Obtain a remote sensing image of the forest area to be measured; the remote sensing image is an image obtained by taking partitioned photos of the forest area to be measured through remote sensing technology; Obtain the elevation values of each monitoring station in the forest area to be measured and the gray values of each monitoring station at the corresponding positions in the remote sensing image, and establish an elevation-gray relationship model based on the elevation values and gray values of each monitoring station; For each partition of the forest area to be measured, determine the vegetation coverage density of the partition based on the gray values of each pixel point in the remote sensing image corresponding to the partition; Determine the theoretical elevation value corresponding to the gray value of each pixel point in the remote sensing image corresponding to the partition based on the elevation-gray relationship model, and correct the measured elevation value based on the vegetation coverage density and the theoretical elevation value to obtain a weighted elevation value; Determine the seasonal weight based on the temperature of the forest area to be measured, and determine the target elevation value of the partition based on the seasonal weight and the weighted elevation value.

[0005] Correspondingly, the present invention also provides a forest terrain remote sensing mapping system for smart forestry, specifically including: An image acquisition module, configured to acquire a remote sensing image of the forest area to be measured; the remote sensing image is an image obtained by partitioning and photographing the forest area to be measured through remote sensing technology; A model establishment module, configured to obtain the elevation values of each monitoring station in the forest area to be measured and the gray values at the corresponding positions of each monitoring station in the remote sensing image, and establish an elevation-gray relationship model based on the elevation values and the gray values of each monitoring station; A data correction module, configured to, for each partition of the forest area to be measured, determine the vegetation coverage density of the partition based on the gray value of each pixel point in the remote sensing image corresponding to the partition; The data correction module is further configured to determine the theoretical elevation value corresponding to the gray value of each pixel point in the remote sensing image corresponding to the partition based on the elevation-gray relationship model, and correct the measured elevation value based on the vegetation coverage density and the theoretical elevation value to obtain a weighted elevation value; The data correction module is further configured to determine the seasonal weight based on the temperature of the forest area to be measured, and determine the target elevation value of the partition based on the seasonal weight and the weighted elevation value.

[0006] The present invention may have the following partial or all beneficial effects: In the forest terrain remote sensing mapping method for smart forestry provided by the present invention, a remote sensing image of the forest area to be measured is acquired; the above-mentioned remote sensing image is an image obtained by taking partitioned photos of the forest area to be measured through remote sensing technology; the elevation values of each monitoring station in the forest area to be measured and the gray values at the corresponding positions of each monitoring station in the remote sensing image are acquired, and an elevation-gray relationship model is established based on the elevation values and gray values of each monitoring station; for each partition of the forest area to be measured, the vegetation coverage density of the partition is determined based on the gray values of each pixel point in the remote sensing image corresponding to the partition; based on the elevation-gray relationship model, the theoretical elevation values corresponding to the gray values of each pixel point in the remote sensing image corresponding to the partition are determined, and the measured elevation values are corrected based on the vegetation coverage density and the theoretical elevation values to obtain weighted elevation values; the seasonal weight is determined based on the temperature in the forest area to be measured, and the target elevation value of the partition is determined based on the seasonal weight and the weighted elevation value. The present invention establishes an elevation-gray relationship model based on the elevation values of each monitoring station and the gray values at their corresponding positions in the remote sensing image, so that the theoretical elevation values corresponding to the gray values of each pixel point can be obtained based on this elevation-gray relationship model; in addition, after obtaining the theoretical elevation values, the present invention also takes into account the influence of the vegetation coverage rate and temperature on the elevation data, preliminarily corrects the measured elevation values through the vegetation coverage rate and the theoretical elevation values to obtain weighted elevation values, and further adjusts the weighted elevation values based on the seasonal weight corresponding to the temperature to obtain the target elevation value, solving the problem that in the process of forest remote sensing mapping, due to the thick tree canopies in some areas, it is difficult for optical remote sensing equipment to penetrate the tree canopies to reach the ground, resulting in errors in the detection of surface elevation information, improving the accuracy of elevation data, and thus helping to obtain more accurate topographic mapping results.

[0007] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. Brief Description of the Drawings

[0008] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0009] Figure 1 Shows a flowchart of a forest terrain remote sensing mapping method for smart forestry according to an exemplary embodiment of the present disclosure; Figure 2 Shows a schematic diagram of a forest terrain remote sensing mapping system for smart forestry according to an exemplary embodiment of the present disclosure. Detailed Embodiments

[0010] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following specifically describes, with reference to the accompanying drawings and preferred embodiments, the specific implementation manner, structure, features and effects of the forest terrain remote sensing mapping method for intelligent forestry proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0011] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0012] The following specifically describes the specific solution of the forest terrain remote sensing mapping method for intelligent forestry provided by the present invention with reference to the accompanying drawings.

[0013] Please refer to Figure 1 , which shows the method flow chart of the forest terrain remote sensing mapping method for intelligent forestry provided by an embodiment of the present invention. As Figure 1 shown, the forest terrain remote sensing mapping method for intelligent forestry specifically includes the following steps: S110: Obtain the remote sensing image of the forest area to be measured; the remote sensing image is an image obtained by taking sectional photos of the forest area to be measured through remote sensing technology; S120: Obtain the elevation values of each monitoring station in the forest area to be measured and the gray values at the corresponding positions of each monitoring station in the remote sensing image, and establish an elevation-gray relationship model based on the elevation values and gray values of each monitoring station; S130: For each partition of the forest area to be measured, determine the vegetation coverage density of the partition based on the gray values of each pixel point in the remote sensing image corresponding to the partition; S140: Determine the theoretical elevation values corresponding to the gray values of each pixel point in the remote sensing image corresponding to the partition based on the elevation-gray relationship model, and correct the measured elevation values based on the vegetation coverage density and the theoretical elevation values to obtain weighted elevation values; S150: Determine the seasonal weight based on the temperature of the forest area to be measured, and determine the target elevation value of the partition based on the seasonal weight and the weighted elevation value.

[0014] The present invention establishes an elevation - grayscale relationship model based on the elevation values of each monitoring station and the grayscale values at the corresponding positions in the remote - sensing image, so that the theoretical elevation value corresponding to the grayscale value of each pixel can be obtained based on this elevation - grayscale relationship model. In addition, after obtaining the theoretical elevation value, the present invention also takes into account the influence of vegetation coverage rate and temperature on elevation data. The measured elevation value is preliminarily corrected through the vegetation coverage rate and the theoretical elevation value to obtain a weighted elevation value, and the weighted elevation value is further adjusted based on the seasonal weight corresponding to the temperature to obtain the target elevation value. This solves the problem that in the process of forest remote - sensing mapping, due to the thick tree canopies in some areas, it is difficult for optical remote - sensing equipment to penetrate the tree canopies to reach the ground, resulting in errors in the detection of surface elevation information, improves the accuracy of elevation data, and thus helps to obtain more accurate topographic mapping results.

[0015] Next, each step of the above - mentioned forest terrain remote - sensing mapping method for smart forestry will be described in detail: In step S110, a remote - sensing image of the forest area to be measured is obtained; the remote - sensing image is an image obtained by taking sub - area photos of the forest area to be measured through remote - sensing technology.

[0016] In the embodiment of the present application, the above - mentioned forest area to be measured is any forest area. The forest terrain remote - sensing mapping method provided by the embodiment of the present application is used to correct the elevation value of the forest area to be measured obtained through remote - sensing technology, so as to improve the accuracy of terrain measurement and control of the forest area to be measured.

[0017] In the embodiment of the present application, the above - mentioned remote - sensing image is an image that reflects the objects and phenomena on the earth's surface corresponding to the forest area to be measured obtained through remote - sensing technology. Specifically, since the area of the above - mentioned forest area to be measured is generally relatively large and the resolution of the shooting equipment on the aerial platform is limited, in order to ensure the quality of the remote - sensing image, the embodiment of the present application can take sub - area photos of the forest area by setting appropriate resolution and altitude. Exemplarily, in the process of the above - mentioned sub - area shooting, the forest area can be divided into several relatively independent areas according to factors such as the natural boundaries (such as rivers, ridge lines) of the forest area to be measured, the distribution of vegetation types, and management requirements (such as different logging areas, protected areas); the basis for choosing the above - mentioned altitude can be as follows: for areas with large area and relatively flat terrain, a higher flight altitude can be selected to obtain a wider field of view and a larger - scale image. For areas with complex terrain or special research needs, the flight altitude is reduced to obtain clearer details; the basis for choosing the above - mentioned resolution can be as follows: the resolution required for macro - monitoring areas is lower than that for areas where the types of vegetation and the health status of trees need to be analyzed in detail. The former only needs to clearly show the general range of the forest area and the overall situation of vegetation coverage, while the latter needs to distinguish the morphology of individual trees and some subtle features.

[0018] Specifically, the above-mentioned acquisition of the remote sensing image of the forest area to be measured can be realized as follows: By using lidar technology, by emitting laser beams and measuring the time when they are reflected back, three-dimensional information of the ground and objects in the forest area to be measured is obtained, and high-precision point cloud data is generated; the point cloud data is processed by denoising, resampling, triangular meshing, surface fitting, projection conversion, rasterization, etc., and is converted into a remote sensing image of the forest area to be measured.

[0019] In step S120, the elevation values of each monitoring station in the forest area to be measured and the gray values at the corresponding positions of each monitoring station in the remote sensing image are obtained, and an elevation-gray relationship model is established based on the elevation values and gray values of each monitoring station.

[0020] In the embodiment of the present application, the above-mentioned elevation value is the value of elevation data. Among them, the above-mentioned elevation data is an important data for describing the terrain undulation of the earth's surface, which is the vertical distance from a ground point to the geoid or other reference planes.

[0021] In the embodiment of the present application, the above-mentioned gray value is a numerical value representing the gray level of the remote sensing image obtained through step S110.

[0022] In the embodiment of the present application, the height difference and vegetation coverage degree at different positions in the forest area to be measured will be reflected in terms of gray scale. For example, in low-altitude areas, the climate is usually warm and humid, which is suitable for the growth of various types of forests. Therefore, the vegetation grows lushly, and the reflectivity of the forest is usually low, and the gray value of the remote sensing image is darker. As the altitude increases, the temperature decreases, the air is thin, and the reflectivity of the ground surface is relatively high, and the gray value of the remote sensing image is usually brighter. Therefore, the elevation value can be predicted based on the relationship between the gray value and the elevation value. Specifically, the above-mentioned elevation-gray relationship model can be established to realize the prediction of the elevation value based on the gray value.

[0023] Exemplarily, the above-mentioned elevation-gray relationship model can be established by setting multiple monitoring stations in the forest area to be measured and through the relationship between the elevation values of each obtained monitoring station and the gray values at the corresponding positions in the remote sensing image. Specifically, this process can be realized as follows: Obtain the elevation values of each monitoring station, and arrange the elevation values in a preset order to obtain an elevation data sequence; determine the gray values at the corresponding positions of each monitoring station in the remote sensing image to obtain a gray data sequence corresponding to the elevation data sequence; determine a regression equation based on the elevation data sequence and the gray data sequence to obtain an elevation-gray relationship model.

[0024] The above monitoring stations need to be set at different height positions as much as possible and ensure clear weather during monitoring. In addition, it is also necessary to ensure that the monitoring stations are distributed at different positions in the forest area to be measured, and avoid centralized setting to reduce measurement errors. Exemplarily, the setting of the above monitoring stations can be achieved as follows: divide the entire forest area to be measured into multiple blocks, and set monitoring stations at the lowest and highest positions in each block. After setting up the monitoring stations, further, the elevation values of each monitoring station can be obtained through the following method: measure the height difference Δh between the monitoring station and the known reference point at the monitoring station with a level (theodolite) and a leveling staff; take the sum H = H0 + Δh of the elevation value H0 of the known reference point and the above height difference as the elevation value of the monitoring station.

[0025] The above-mentioned obtaining of the elevation values of each monitoring station and arranging the elevation values in a preset order to obtain an elevation data sequence can be specifically achieved as follows: after obtaining the elevation values of each monitoring station, arrange the elevation data of each monitoring station from low to high to obtain an elevation data sequence. 。

[0026] The above-mentioned determination of the gray values at the corresponding positions of each monitoring station in the remote sensing image to obtain the gray data sequence corresponding to the elevation data sequence can be specifically achieved as follows: mark the position of the monitoring station in the remote sensing image and obtain the gray value of the remote sensing image at that position, and arrange the gray values according to the order of the monitoring stations corresponding to the elevation values in the above elevation data series to obtain the gray data sequence corresponding to the elevation data sequence. The corresponding gray data sequence 。

[0027] The above-mentioned determination of the regression equation based on the elevation data sequence and the gray data sequence to obtain the elevation-gray relationship model can be achieved as follows: calculate the slope and intercept of the regression equation by the least squares method based on the elevation data sequence and the gray data sequence; determine the local slope of the regression equation at different positions based on the elevation data sequence and the gray data sequence, determine the non-linearity of the elevation data sequence and the gray data sequence based on the local slope, and take the absolute value of the non-linearity as the absolute value of the error term of the regression equation; determine the positive and negative of the error term based on the data distribution of the regression equation; substitute the slope, intercept and error term into the regression equation to obtain the elevation-gray relationship model.

[0028] Taking the above elevation data sequence and the above gray data sequence as an example, the above-mentioned calculation of the slope and intercept by the least squares method based on the elevation data sequence and the gray data sequence to obtain the corresponding regression equation is specifically as follows:

[0029] Among them, the above is the elevation value, is the gray value, is the slope of the regression equation calculated based on the above elevation data sequence and the above grayscale data sequence is the intercept of the regression equation calculated based on the above elevation data sequence and the above grayscale data sequence is the error term is the error term is the error term

[0030] In addition, due to factors such as the randomness of the monitoring site location and the irregular vegetation coverage at the same elevation, although there is a positive correlation between the above elevation values and the corresponding grayscale values, there may be a non-linear situation. To improve the accuracy of the above elevation-grayscale relationship model, the error term is introduced in the embodiments of the present application to correct the regression equation, so that the regression equation can better fit the relationship between grayscale and elevation

[0031] Exemplarily, the local slope of the regression equation determined based on the elevation data sequence and the grayscale data sequence at different positions, and determining the non-linearity of the elevation data sequence and the grayscale data sequence based on the local slope can achieve the following: for adjacent elevation data in the elevation data sequence, calculate the first difference between the adjacent elevation data, and the second difference between the corresponding adjacent grayscale data in the elevation data sequence; calculate the ratio of the first difference and the second difference to obtain the local slope; calculate the third difference between the adjacent local slopes, and determine the non-linearity based on the sum of the third differences and the number of monitoring stations

[0032] Specifically, taking the above elevation data sequence and the above grayscale data sequence as an example, the specific calculation formula of the above local slope can be as follows

[0033] where is the local slope of the above elevation data sequence and the above grayscale data sequence at the th position and are the th elevation value and the (i - 1)th elevation value in the above elevation data sequence respectively, and is the above first difference and are the th grayscale value and the (i - 1)th grayscale value in the above grayscale data sequence respectively, and

[0034] is the above second difference After obtaining the above local slope After that, further, based on the above elevation data sequence and the above grayscale data sequence The non - linearity between the two data sequences can be determined according to the local slopes at different positions. Specifically, the non - linearity can be calculated by the following formula:

[0035] Wherein, is the non - linearity between the above elevation data sequence and the above grayscale data sequence ; is the number of monitoring stations; By comparing all the local slopes of the above elevation data sequence and the grayscale data sequence at each position to measure the linear degree between the two. If they are linearly correlated, all the corresponding local slopes of the above elevation data sequence and the grayscale data sequence are the same, and the value of U tends to 0; if they are non - linearly correlated, the greater the value of U, the greater the non - linear degree between the above elevation data sequence and the grayscale data sequence .

[0036] The non - linear relationship between the above elevation data sequence and the grayscale data sequence results in that not all ( ) data points pass through the fitting curve corresponding to the regression equation. In order to make more data points fall on the fitting curve, the fitting curve can be moved in the direction where there are more data points conforming to the regression equation to improve the fitting effect of the regression equation. Therefore, after calculating the non - linearity (i.e., the magnitude of the error term) by the above method, the embodiments of the present application also need to determine the positive or negative of the error term according to the data distribution to determine the moving direction of the fitting curve. Exemplarily, this process can be implemented as follows: determine the first data points that do not pass through the fitting curve and are located in the area above the fitting curve corresponding to the regression equation, and the second data points that do not pass through the fitting curve and are located in the area below the fitting curve; if the number of the first data points is greater than the number of the second data points, the error term takes a negative value; if the number of the first data points is less than the number of the second data points, the error term takes a positive value. Specifically, the positive or negative of the error term can be determined by the following formula:

[0037] Wherein, represents the scale difference of the data points that do not pass through the regression equation, and is used to describe the deviation scale of the data points from the trend or model represented by the regression equation; is the number of data points in the region above the fitting curve corresponding to the regression equation (i.e., the number of the above first data points); is the number of data points in the region below the fitting curve corresponding to the regression equation (i.e., the number of the above second data points); If it is positive, it proves that the regression equation should be moved downward and the error term should be less than 0; If it is negative, it proves that the regression equation should be moved upward and the error term should be greater than 0.

[0038] Based on the above elevation data sequence and grayscale data sequence calculate the slope and intercept of the regression equation, and after determining the magnitude and sign of the error term through the above process, substitute the calculated slope intercept and error term into the above regression equation: a more accurate elevation - grayscale relationship model can be obtained.

[0039] In step S130, for each partition of the forest area to be measured, determine the vegetation coverage density of the partition based on the grayscale values of each pixel point in the remote sensing image corresponding to the partition.

[0040] In the embodiments of the present application, the above vegetation coverage density is used to measure the degree or density of vegetation coverage in each partition area of the forest area to be measured. Since the vegetation coverage density will affect the distribution of grayscale in the remote sensing image. For example, for an area with a high vegetation coverage density, the leaves fully cover the ground surface, and the degree of ground surface exposure is small, and the corresponding remote sensing image has a deep and uniform grayscale distribution; on the contrary, for an area with a low vegetation coverage density, the degree of ground surface exposure is large, and the corresponding remote sensing image has a shallow and uniform grayscale distribution. Therefore, the vegetation coverage density at the corresponding position can be determined based on the grayscale values of each pixel point in the remote sensing image.

[0041] Exemplarily, the above determination of the vegetation coverage density of each partition of the forest area to be measured based on the grayscale values of each pixel point in the remote sensing image corresponding to the partition can be achieved as follows: For each partition of the forest area to be measured, perform clustering processing on the grayscale values of each pixel point in the remote sensing image corresponding to the partition to obtain multiple result cluster regions; determine whether each result cluster region is a bare ground surface region based on the average grayscale value of each result cluster region; determine the vegetation coverage density based on the area of the bare ground surface region and the grayscale values of each pixel point in the non - bare ground surface region.

[0042] Specifically, for each partition of the forest area to be measured, based on the relationship between the above-mentioned vegetation coverage density and the gray-scale distribution in the remote sensing image, clustering can be performed on the gray-scale in each partition through a clustering method to obtain multiple result cluster regions, and the gray-scales within the same result cluster region are more similar.

[0043] After obtaining multiple result cluster regions through clustering operations, exemplarily, the determination of whether each result cluster region is a bare ground area based on the average gray-scale value of each result cluster region can be achieved as follows: for each result cluster region, determine the first average gray-scale value of the current result cluster region and the second average gray-scale value of other result cluster regions except the current result cluster region; determine the surface bare value of the current result cluster region based on the first average gray-scale value, the second average gray-scale value, the near-infrared band reflectance of the current result cluster region, and the visible light band reflectance of the current result cluster region; if the surface bare value is less than the preset threshold, determine that the current result cluster region is a non-bare ground area, otherwise, determine that the current result cluster region is a bare ground area.

[0044] Specifically, the above-mentioned surface bare value can be determined by the following formula:

[0045] where, used to describe the possibility that a certain result cluster region is a bare ground area (i.e., the above-mentioned surface bare value); is the average gray-scale value of this result cluster region (i.e., the above-mentioned first average gray-scale value), is the average gray-scale value of any other result cluster region except this result cluster region (i.e., the above-mentioned second average gray-scale value); is the number of result cluster regions obtained by clustering. is the result cluster region 's near-infrared band reflectance, is the result cluster region 's visible light band reflectance.

[0046] In the above formula, the above-mentioned represents the average gray-scale difference between the result cluster region and other result cluster regions. Since the overall gray-scale of the bare ground is lighter than that of the vegetation coverage (the larger the gray-scale value), so the larger the value, the higher the possibility that the result cluster region is a bare ground area; the above-mentioned represents the result cluster region Regarding the light reflection characteristics, since the reflectance of vegetation in the red light (visible light) band and the near-infrared band is relatively higher than that of the bare surface area, therefore The larger the value of , the higher the possibility that the corresponding result cluster area

[0047] After determining the surface bare value of any result cluster area through the above formula, further, the embodiments of the present disclosure can also normalize the calculated surface bare value to the range of [0, 1], and determine whether the corresponding result cluster area is a bare surface area by setting a preset threshold. Taking the preset threshold as 0.85 as an example, then The result cluster area of

[0048] After determining the bare surface area in any sub-region of the forest area to be measured, exemplarily, the determination of the vegetation coverage density based on the area of the bare surface area and the gray values of each pixel point in the non-bare surface area can be realized as follows: determining the area proportion of the non-bare surface area in the sub-region based on the area of the sub-region and the area of the bare surface area; calculating the fourth difference between the gray values of any two pixel points in the non-bare surface area, and determining the uniformity of the gray values of the non-bare surface area based on the sum of the fourth differences; determining the vegetation coverage density based on the area proportion of the non-bare surface area in the sub-region and the uniformity of the gray values of the non-bare surface area.

[0049] Specifically, the vegetation coverage density of any sub-region in the above forest area to be measured can be determined by the following formula:

[0050] Wherein, represents the vegetation coverage density of the current sub-region; is the actual area of the current sub-region, is the area of the bare surface in the current sub-region; represents the gray value of the a-th pixel point in the non-bare surface area of the current sub-region; represents the gray value of the b-th pixel point in the non-bare surface area of the current sub-region; A represents the total number of pixel points in the non-bare surface area of the current sub-region.

[0051] In the above formula, the above represents the area proportion of the non-bare surface in the current sub-region. The larger the value of this area proportion, it proves that the area proportion of the bare surface is smaller, and the vegetation coverage density of the current sub-region is high; the above represents the uniformity of the gray values of all pixel points in the non-bare surface area of the current sub-region, The smaller the value is, the more uniform the grayscale of all pixels in the non-bare surface area in the current partition is, and the greater the corresponding vegetation coverage density is; is a normalization function used to convert the calculated vegetation coverage density Normalized to the range [0,1].

[0052] In step S140, the theoretical elevation value corresponding to the grayscale value of each pixel in the remote sensing image corresponding to the partition is determined based on the elevation-grayscale relationship model, and the measured elevation value is corrected based on the vegetation coverage density and the theoretical elevation value to obtain a weighted elevation value.

[0053] In an embodiment of the present application, the above-mentioned theoretical elevation value is the elevation value obtained by inputting the grayscale value of any pixel point in the remote sensing image into the above-mentioned elevation-grayscale relationship model; the above-mentioned measured elevation value is the elevation value of the corresponding position of the forest area to be measured measured by remote sensing technology; the above-mentioned weighted elevation value is the elevation value obtained by correcting the measured elevation value based on the above-mentioned vegetation coverage density and the theoretical elevation value.

[0054] In the embodiment of the present application, since vegetation coverage will affect the accuracy of the elevation data measurement results of the forest area to be measured, that is, the greater the vegetation coverage density of a certain partition, it proves that the elevation value of the partition is more seriously affected by the vegetation coverage, and the corresponding elevation data deviation is greater, which requires a greater degree of correction. Therefore, when predicting the elevation value through the grayscale value, the vegetation coverage density can be used as an influencing factor. The grayscale conditions of the remote sensing images in different partitions of the forest area to be measured are compared through the above-mentioned elevation-grayscale model, and the vegetation coverage density of different partitions is calculated at the same time. Thereby, the elevation correction factor is calculated according to the grayscale difference of different partitions and the vegetation coverage density, and the measured elevation value is corrected by the calculated elevation correction factor.

[0055] Exemplarily, the above-mentioned determination of the theoretical elevation value corresponding to the grayscale value of each pixel in the remote sensing image corresponding to the partition based on the elevation-grayscale relationship model, and correcting the measured elevation value based on the vegetation coverage density and the theoretical elevation value to obtain the weighted elevation value can be achieved as follows: inputting the grayscale value of each pixel in the remote sensing image corresponding to the partition into the elevation-grayscale relationship model, and determining the theoretical elevation value corresponding to the input grayscale value based on the regression equation of the elevation-grayscale relationship model; obtaining the measured elevation value obtained by measurement corresponding to the theoretical elevation value; determining the weights of the theoretical elevation value and the measured elevation value based on the vegetation coverage density, and determining the weighted elevation value based on the theoretical elevation value, the measured elevation value, and the weights of the theoretical elevation value and the measured elevation value.

[0056] Specifically, the above weighted elevation value can be determined by the following formula:

[0057] in, For a certain pixel in the current partition The weighted elevation value at the corresponding position in the forest area to be measured, is the measured elevation value obtained by measuring at the corresponding position of the pixel through lidar technology in the forest area to be measured, is the theoretical elevation value of the pixel obtained through the elevation-gray relationship model at the corresponding position in the forest area to be measured; is the vegetation coverage density of the current partition.

[0058] In the above formula, since the vegetation coverage density will affect the accuracy of the measured elevation value the greater the vegetation coverage density, the smaller the confidence level needs to be set for the measured elevation value, and the greater the confidence level needs to be set for the theoretical elevation value. That is, using the vegetation coverage density as the weight to set different weights for and thereby weightedly correcting the elevation value at the corresponding position in the forest area to be measured for any pixel, and improving the accuracy of the elevation data of the forest area to be measured.

[0059] In step S150, determine the seasonal weight based on the temperature of the forest area to be measured, and determine the target elevation value of the partition based on the seasonal weight and the weighted elevation value.

[0060] In the embodiments of the present application, the weighted elevation value determined through the above steps S110 to S140 is the elevation value calculated based on the remote sensing image in which the trees and the ground both maintain a natural state under ideal weather. However, in the actual scenario, there may be significant differences in the vegetation coverage rate of the forest area to be measured in different seasons. Therefore, the method of the above steps S110 to S140 needs to be further corrected for the weighted elevation value corrected based on the vegetation coverage density without specifying the measurement season, so as to further improve the accuracy of the finally obtained target elevation value. The above seasonal weight is used for the further correction of the weighted elevation value.

[0061] Since the influence of seasons on vegetation coverage is mainly reflected in temperature, the above seasonal temperature can be set based on temperature. Exemplarily, the above determining the seasonal weight based on the temperature of the forest area to be measured and determining the target elevation value of the partition based on the seasonal weight and the weighted elevation value can be implemented as follows: obtain the average temperature of the last month of each quarter in the forest area to be measured, and determine the seasonal weight corresponding to each quarter based on the obtained average temperature; correct the weighted elevation value of the corresponding quarter of the forest area to be measured based on the seasonal weight to obtain the target elevation value.

[0062] Specifically, the average temperature of the last month of each quarter in the past year of the forest area to be measured can be obtained , and based on the average temperature of the last month of each quarter obtained determine the seasonal weights as follows:

[0063] Among them, is the seasonal weight of the forest area to be measured in the th quarter; is the average temperature of the last month of the th quarter; is the average temperature of the last month of the th quarter.

[0064] Furthermore, the weighted elevation value of the corresponding quarter of the forest area to be measured can be corrected based on the seasonal weight as follows:

[0065] Among them, is the target elevation value of a certain pixel point at the corresponding position in any partition of the forest area to be measured; is the weighted elevation value of the pixel point at the corresponding position in the th quarter; is the seasonal weight of the forest area to be measured in the th quarter.

[0066] In the embodiment of the present application, after obtaining the target elevation values at various positions of the forest area to be measured through the above process, further, the embodiment of the present application can also perform the following processing on the obtained target elevation values to accurately monitor, manage, and protect forestry resources through the target elevation values: represent the target elevation values at different positions of the forest area to be measured obtained through the above steps in the form of a grid, and each grid unit (pixel) contains a height value; organize the grid units corresponding to the target elevation values into a regular grid. Exemplarily, each grid can represent an area of 1m×1m, and generate a corresponding digital elevation model; through each target elevation point in the digital elevation model, generate corresponding coordinate points (X, Y, Z) in three-dimensional space, and connect the generated coordinate points through modeling software to obtain a corresponding three-dimensional grid model, and assist in the accurate monitoring, management, and protection of forestry resources based on this three-dimensional grid model; among them, the above three-dimensional grid model can represent the terrain undulation of the forest area to be measured.

[0067] The present invention establishes an elevation - grayscale relationship model based on the elevation values of each monitoring station and the grayscale values at the corresponding positions in the remote sensing image, so that the theoretical elevation value corresponding to the grayscale value of each pixel can be obtained based on this elevation - grayscale relationship model; in addition, after obtaining the theoretical elevation value, the present invention also takes into account the influence of vegetation coverage rate and temperature on elevation data, preliminarily corrects the measured elevation value through the vegetation coverage rate and the theoretical elevation value to obtain a weighted elevation value, and further adjusts the weighted elevation value based on the seasonal weight corresponding to the temperature to obtain the target elevation value, solving the problem that in the process of forest remote sensing mapping, due to the thick tree canopies in some areas, it is difficult for optical remote sensing equipment to penetrate the tree canopies to reach the ground, resulting in errors in the detection of surface elevation information, improving the accuracy of elevation data, and thus contributing to obtaining more accurate topographic mapping results.

[0068] The above mainly introduces the solution provided by the embodiments of the present invention from the perspective of methods. To implement the above functions, it includes the corresponding hardware structure and / or software module for executing each function. Those skilled in the art should easily realize that, combining the units and algorithm steps of each example described in the embodiments disclosed herein, the present invention can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the way of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0069] Embodiments of the present disclosure provide a forest terrain remote sensing mapping system for smart forestry. Refer to Figure 2 As shown, the forest terrain remote sensing mapping system 400 for smart forestry may include an image acquisition module 410, a model establishment module 420, and a data correction module 430, where: The image acquisition module 410 can be used to acquire remote sensing images of the forest area to be measured; the above remote sensing images are images obtained by taking sectional photos of the forest area to be measured through remote sensing technology; The model establishment module 420 can be used to obtain the elevation values of each monitoring station in the forest area to be measured and the grayscale values at the corresponding positions of each monitoring station in the remote sensing image, and establish an elevation - grayscale relationship model based on the elevation values and grayscale values of each monitoring station; The data correction module 430 can be used to determine the vegetation coverage density of each partition in the forest area to be measured based on the grayscale values of each pixel in the remote sensing image corresponding to the partition; The data correction module 430 is also used to determine the theoretical elevation value corresponding to the grayscale value of each pixel in the remote sensing image corresponding to the partition based on the elevation - grayscale relationship model, and correct the measured elevation value based on the vegetation coverage density and the theoretical elevation value to obtain a weighted elevation value; The data correction module 430 is further configured to determine a seasonal weight based on the temperature of the forest area to be measured, and determine a target elevation value of the partition based on the seasonal weight and the weighted elevation value.

[0070] In an embodiment of the present application, the above model establishment module is specifically configured to: obtain the elevation values of each monitoring station, arrange the elevation values in a preset order to obtain an elevation data sequence; determine the gray values at the corresponding positions of each monitoring station in the remote sensing image to obtain a gray data sequence corresponding to the elevation data sequence; determine a regression equation based on the elevation data sequence and the gray data sequence to obtain an elevation-gray relationship model.

[0071] In an embodiment of the present application, the above model establishment module is specifically configured to: calculate the slope and intercept of the regression equation by the least squares method based on the elevation data sequence and the gray data sequence; determine the local slope of the regression equation at different positions based on the elevation data sequence and the gray data sequence, determine the non-linearity of the elevation data sequence and the gray data sequence based on the local slope, and use the absolute value of the non-linearity as the absolute value of the error term of the regression equation; determine the positive and negative of the error term based on the data distribution of the regression equation; substitute the slope, intercept and error term into the regression equation to obtain an elevation-gray relationship model.

[0072] In an embodiment of the present application, the above model establishment module is specifically configured to: for adjacent elevation data in the elevation data sequence, calculate a first difference between the adjacent elevation data and a second difference between the adjacent gray data corresponding to the elevation data sequence; calculate the ratio of the first difference and the second difference to obtain a local slope; calculate a third difference between adjacent local slopes, and determine the non-linearity based on the sum of the third differences and the number of monitoring stations.

[0073] In an embodiment of the present application, the above data correction module is specifically configured to: for each partition of the forest area to be measured, perform clustering processing on the gray values of each pixel point in the remote sensing image corresponding to the partition to obtain a plurality of result cluster regions; determine whether each result cluster region is an exposed surface area based on the average gray value of each result cluster region; determine the vegetation coverage density based on the area of the exposed surface area and the gray values of each pixel point in the non-exposed surface area.

[0074] In an embodiment of the present application, the above data correction module is specifically configured to: for each result cluster region, determine a first average gray value of the current result cluster region and a second average gray value of other result cluster regions except the current result cluster region; determine the surface exposure value of the current result cluster region based on the first average gray value, the second average gray value, the near-infrared band reflectance of the current result cluster region and the visible light band reflectance of the current result cluster region; if the surface exposure value is less than a preset threshold, determine that the current result cluster region is a non-exposed surface area, otherwise, determine that the current result cluster region is an exposed surface area.

[0075] In the embodiment of the present application, the above data correction module is specifically configured to: determine the area proportion of the non-exposed surface area in the partition based on the area of the partition and the area of the exposed surface area; calculate the fourth difference between the gray values of any two pixel points in the non-exposed surface area, and determine the uniformity of the gray values of the non-exposed surface area based on the sum of the fourth differences; determine the vegetation coverage density based on the area proportion of the non-exposed surface area in the partition and the uniformity of the gray values of the non-exposed surface area.

[0076] In the embodiment of the present application, the above data correction module is specifically configured to: input the gray values of each pixel point in the remote sensing image corresponding to the partition into the elevation-gray value relationship model, and determine the corresponding theoretical elevation value based on the regression equation; obtain the measured elevation value obtained by measurement corresponding to the theoretical elevation value; determine the weights of the theoretical elevation value and the measured elevation value based on the vegetation coverage density, and determine the weighted elevation value based on the theoretical elevation value, the measured elevation value, and the weights of the theoretical elevation value and the measured elevation value. In the embodiment of the present application, the above data correction module is specifically configured to: obtain the average temperature of the forest area to be measured in the last month of each quarter, and determine the seasonal weight corresponding to each quarter based on the average temperature; correct the weighted elevation value of the corresponding quarter of the forest area to be measured based on the seasonal weight to obtain the target elevation value.

[0077] The embodiment of the present invention can divide the function modules of the forest terrain remote sensing mapping system for intelligent forestry according to the above method examples. For example, each function can correspond to each function module, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of a software function module. It should be noted that the division of modules in the embodiment of the present invention is illustrative, only a logical function division, and there may be other division methods in actual implementation.

[0078] In addition, the specific implementation details of the above forest terrain remote sensing mapping system for intelligent forestry have been described in detail at the corresponding positions of the forest terrain remote sensing mapping method for intelligent forestry, so they will not be repeated here.

[0079] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0080] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A forest terrain remote sensing mapping method for smart forestry, characterized in that, The method includes: Obtaining a remote sensing image of the forest area to be measured; the remote sensing image is an image obtained by partitioning and photographing the forest area to be measured through remote sensing technology; Obtaining the elevation values of each monitoring station in the forest area to be measured and the gray values at the corresponding positions of each monitoring station in the remote sensing image, and establishing an elevation-gray relationship model based on the elevation values and the gray values of each monitoring station; For each partition of the forest area to be measured, determining the vegetation coverage density of the partition based on the gray values of each pixel point in the remote sensing image corresponding to the partition; Determining the theoretical elevation values corresponding to the gray values of each pixel point in the remote sensing image corresponding to the partition based on the elevation-gray relationship model, and correcting the measured elevation values based on the vegetation coverage density and the theoretical elevation values to obtain weighted elevation values; Determining a season weight based on the temperature in the forest area to be measured, and determining the target elevation value of the partition based on the season weight and the weighted elevation value.

2. The forest terrain remote sensing mapping method for smart forestry according to claim 1, wherein The obtaining the elevation values of each monitoring station in the forest area to be measured and the gray values at the corresponding positions of each monitoring station in the remote sensing image, and establishing an elevation-gray relationship model based on the elevation values and the gray values of each monitoring station includes: Obtaining the elevation values of each monitoring station, and arranging the elevation values in a preset order to obtain an elevation data sequence; Determining the gray values at the corresponding positions of each monitoring station in the remote sensing image to obtain a gray data sequence corresponding to the elevation data sequence; Determining a regression equation based on the elevation data sequence and the gray data sequence to obtain the elevation-gray relationship model.

3. The forest terrain remote sensing mapping method for intelligent forestry according to claim 2, characterized in that, Determining a regression equation based on the elevation data sequence and the gray data sequence to obtain the elevation-gray relationship model includes: Calculating the slope and intercept of the regression equation by the least squares method based on the elevation data sequence and the gray data sequence; Determining the local slopes of the regression equation at different positions based on the elevation data sequence and the gray data sequence, determining the non-linearity of the elevation data sequence and the gray data sequence based on the local slopes, and taking the absolute value of the non-linearity as the absolute value of the error term of the regression equation; Determining the positive or negative of the error term based on the data distribution of the regression equation; Substituting the slope, the intercept, and the error term into the regression equation to obtain the elevation-gray relationship model.

4. The forest terrain remote sensing mapping method for intelligent forestry according to claim 3, wherein The determining the local slopes of the regression equation at different positions based on the elevation data sequence and the gray data sequence, and determining the non-linearity of the elevation data sequence and the gray data sequence includes: For adjacent elevation data in the elevation data sequence, calculating a first difference between the adjacent elevation data and a second difference between the adjacent gray data corresponding to the elevation data sequence; Calculating the ratio of the first difference and the second difference to obtain the local slope; Calculating a third difference between adjacent local slopes, and determining the non-linearity based on the sum of the third differences and the number of monitoring stations.

5. The forest terrain remote sensing mapping method for intelligent forestry according to claim 3, characterized in that, For each sub-region of the forest area to be measured, determining the vegetation coverage density of the sub-region based on the gray values of each pixel point in the remote sensing image corresponding to the sub-region includes: For each sub-region of the forest area to be measured, performing clustering processing on the gray values of each pixel point in the remote sensing image corresponding to the sub-region to obtain multiple result cluster regions; Determining whether each result cluster region is an exposed ground area based on the average gray value of each result cluster region; Determining the vegetation coverage density based on the area of the exposed ground area and the gray values of each pixel point in the non-exposed ground area.

6. The forest terrain remote sensing mapping method for smart forestry according to claim 5, characterized in that, The determining whether each result cluster region is an exposed ground area based on the average gray value of each result cluster region includes: For each result cluster region, determining a first average gray value of the current result cluster region and a second average gray value of other result cluster regions except the current result cluster region; Determining the surface exposure value of the current result cluster region based on the first average gray value, the second average gray value, the near-infrared band reflectance of the current result cluster region, and the visible light band reflectance of the current result cluster region; If the surface exposure value is less than a preset threshold, determining that the current result cluster region is the non-exposed ground area; otherwise, determining that the current result cluster region is the exposed ground area.

7. The forest terrain remote sensing mapping method for intelligent forestry according to claim 6, wherein, The determining the vegetation coverage density based on the area of the exposed ground area and the gray values of each pixel point in the non-exposed ground area includes: Determining the area proportion of the non-exposed ground area in the sub-region based on the area of the sub-region and the area of the exposed ground area; Calculating a fourth difference between the gray values of any two pixel points in the non-exposed ground area, and determining the uniformity of the gray values of the non-exposed ground area based on the sum of the fourth differences; Determining the vegetation coverage density based on the area proportion of the non-exposed ground area in the sub-region and the uniformity of the gray values of the non-exposed ground area.

8. The forest terrain remote sensing mapping method for smart forestry according to claim 7, characterized in that, The determining the theoretical elevation value corresponding to the gray value of each pixel point in the remote sensing image corresponding to the sub-region based on the elevation-gray relationship model, and correcting the measured elevation value based on the vegetation coverage density and the theoretical elevation value to obtain a weighted elevation value includes: Inputting the gray values of each pixel point in the remote sensing image corresponding to the sub-region into the elevation-gray relationship model, and determining the corresponding theoretical elevation value based on the regression equation; Obtaining the measured elevation value obtained by measurement corresponding to the theoretical elevation value; Determining the weights of the theoretical elevation value and the measured elevation value based on the vegetation coverage density, and determining the weighted elevation value based on the theoretical elevation value, the measured elevation value, and the weights of the theoretical elevation value and the measured elevation value.

9. The forest terrain remote sensing mapping method for smart forestry according to claim 8, characterized in that, The determining the seasonal weight based on the temperature of the forest area to be measured, and determining the target elevation value of the sub-region based on the seasonal weight and the weighted elevation value includes: Obtain the average temperature of the forest area to be measured in the last month of each quarter, and determine the corresponding seasonal weight for each quarter based on the average temperature; Based on the seasonal weight, correct the weighted elevation value corresponding to the quarter of the forest area to be measured to obtain the target elevation value.

10. A forest terrain remote sensing mapping system for smart forestry, characterized in that, The system includes: An image acquisition module for acquiring a remote sensing image of the forest area to be measured; the remote sensing image is an image obtained by partitioning and photographing the forest area to be measured through remote sensing technology; A model establishment module for obtaining the elevation values of each monitoring station in the forest area to be measured and the gray values at the corresponding positions of each monitoring station in the remote sensing image, and establishing an elevation-gray relationship model based on the elevation values and gray values of each monitoring station; A data correction module for determining the vegetation coverage density of each partition based on the gray values of each pixel point in the remote sensing image corresponding to the partition for each partition of the forest area to be measured; The data correction module is further configured to determine the theoretical elevation value corresponding to the gray value of each pixel point in the remote sensing image corresponding to the partition based on the elevation-gray relationship model, and correct the measured elevation value based on the vegetation coverage density and the theoretical elevation value to obtain a weighted elevation value; The data correction module is further configured to determine a seasonal weight based on the temperature of the forest area to be measured, and determine the target elevation value of the partition based on the seasonal weight and the weighted elevation value.

Images