A land cover remote sensing information acquisition method for complex natural scenes

Abstract

The present invention relates to the technical field of remote sensing image analysis and processing, and in particular to a method for collecting land cover remote sensing information for complex natural scenes. The present invention first obtains all tree clusters, and further obtains the difference separation weight of each environmental indicator in each tree cluster, and adjusts the distance between different tree areas in combination with the difference between the corresponding environmental information under the same environmental indicator, to obtain all sub-tree clusters; finally, the tree areas in the sub-tree clusters are image enhanced and labeled for classification. The present invention first preliminarily clusters the tree areas based on the morphological characteristics of the trees, further analyzes the differentiated performance of the environments in which different trees in the same cluster are located, determines the ability of each environmental indicator to distinguish tree species, and then reclassifies the trees in the clusters, and enhances the tree areas to varying degrees, thereby improving the collection effect of land cover remote sensing information.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image analysis and processing, and in particular to a land cover remote sensing information acquisition method for complex natural scenes. Background Art

[0002] Land cover refers to the various natural and artificial coverings on the Earth's surface. Remote sensing technology can monitor large areas in a short period of time, acquiring land cover remote sensing information. This information is a crucial foundation for land cover change analysis, ecosystem monitoring, and resource management. Analyzing tree species in a region based on land cover remote sensing imagery can help assess the composition of that region's ecosystem, enabling precise management and conservation. Therefore, accurately labeling vegetation species is crucial when collecting land cover remote sensing information.

[0003] Currently, tree species in land cover remote sensing images can be classified based on image segmentation or deep learning. However, in complex natural environments, due to the heterogeneity of natural scenes and terrain changes, the characteristics of land cover in remote sensing images may be very complex. The appearance differences of different tree species or even the same tree species in different environments will affect the classification accuracy. At the same time, different tree species may also show similar external manifestations (such as color depth, texture shape, etc.) in remote sensing images, which may lead to tree misclassification, and thus lead to poor collection of land cover remote sensing information. Summary of the Invention

[0004] In order to solve the technical problem of poor collection effect of land cover remote sensing information, the purpose of the present invention is to provide a land cover remote sensing information collection method for complex natural scenes. The technical solution adopted is as follows:

[0005] Obtain all vegetation areas in the remote sensing image of the area to be measured, and obtain environmental information of each vegetation area under each environmental indicator, as well as all tree areas within the vegetation area; the environmental indicators include at least regional altitude, tree density, and distance to water source;

[0006] In remote sensing images, all tree areas are clustered according to their morphological and color characteristics to obtain all tree clusters. Within the preset neighborhood of each tree area, under each environmental indicator, the surrounding environmental complexity of each tree area is obtained based on the differences in the environmental information of all vegetation areas.

[0007] In each of the tree clusters, the difference separation weight of each environmental indicator is obtained according to the difference in the number of tree areas corresponding to each complexity of the surrounding environment, and the distance between different tree areas is adjusted in combination with the difference between the environmental information corresponding to the same environmental indicator to obtain all sub-tree clusters; in each sub-tree cluster, the image of each tree area is enhanced and labeled and classified according to the environmental information under the environmental indicator corresponding to the maximum difference separation weight of all tree areas in the tree cluster to which it belongs.

[0008] Furthermore, the method for obtaining the tree clusters includes:

[0009] The remote sensing image was converted into CIE color space, and the mean a-axis value of all pixels in each tree area was used as the greenness value; the number of corner points in each tree area in the remote sensing image was used as the crown shape parameter;

[0010] The triplet constructed by the greenness value, crown shape parameter and area size of each tree area is used as the characteristic parameter of each tree area, and based on the distance clustering algorithm, the characteristic parameters corresponding to all tree areas in the remote sensing image are clustered, and the corresponding tree clusters are obtained according to the clustering of the characteristic parameters.

[0011] Furthermore, the preset neighborhood of each tree area is all the vegetation areas bordering the vegetation area to which each tree area belongs.

[0012] Furthermore, the method for obtaining the complexity of the surrounding environment includes:

[0013] In a preset neighborhood of each tree area, the standard deviation of the environmental information of all adjacent vegetation areas under each environmental indicator is used as the surrounding environment complexity of each tree area under the corresponding environmental indicator.

[0014] Furthermore, the method for obtaining the difference separation weight includes:

[0015] In each of the tree clusters, under each environmental indicator, a characteristic curve is fitted with the complexity of the surrounding environment as the horizontal axis parameter and the number of tree areas corresponding to the complexity of the surrounding environment as the vertical axis parameter; and based on the degree of fluctuation of the characteristic curve, the difference separation weight under the corresponding environmental indicator is obtained.

[0016] Furthermore, according to the fluctuation degree of the characteristic curve, a method for obtaining the difference separation weight under the corresponding environmental indicator includes:

[0017] In the characteristic curve, the difference separation weight under the corresponding environmental index is obtained according to the change rate between adjacent extreme value points and the total number of extreme value points.

[0018] Furthermore, the method for obtaining the subtree clusters includes:

[0019] In each of the tree clusters, under each environmental indicator, the difference separation weight is used to weight the difference in environmental information between each tree area and the cluster center, and the weighted sum value under all environmental indicators is used as the corrected distance between each tree area and the cluster center; all tree areas in each of the tree clusters are secondary clustered according to the corrected distance to obtain all sub-tree clusters.

[0020] Furthermore, the method of enhancing each tree region includes:

[0021] In each sub-tree cluster, the mean value of the environmental information under the environmental index corresponding to the maximum difference separation weight in the tree cluster to which it belongs is normalized, and the normalized result is used as the enhancement coefficient corresponding to all tree areas in the corresponding sub-tree cluster; the corresponding tree area is enhanced using the enhancement coefficient.

[0022] Furthermore, the method for obtaining the tree area includes:

[0023] Edge detection is performed on each vegetation area to obtain all closed edges, and the area enclosed by each closed edge is regarded as a suspected tree area; the average area of ​​all suspected tree areas is used as a reference area, and the suspected tree areas smaller than the reference area are regarded as tree areas.

[0024] The present invention has the following beneficial effects:

[0025] The present invention first obtains all vegetation areas in the remote sensing image of the area to be measured, and obtains the environmental information of each vegetation area under each environmental indicator, as well as all tree areas within it, to provide an analysis basis; then, in the remote sensing image, all tree areas are preliminarily clustered according to the morphological characteristics and color characteristics of each tree area to obtain all tree clusters; further, the surrounding environmental complexity of each tree area is obtained. The surrounding environmental complexity provides a certain reference for the growth geographical environment performance of each tree area under each environmental indicator, and subsequently, under each environmental indicator, the distribution of tree areas in the same cluster under similar surrounding environmental complexity is combined to evaluate the impact of each environmental indicator on tree growth. Long impact is used to prepare for obtaining difference separation weights; then, in each tree cluster, according to the difference in the number of tree areas corresponding to each surrounding environment complexity, a difference separation weight reflecting the ability of each environmental indicator to help distinguish different tree species in each tree cluster is obtained, and combined with the difference between the environmental information corresponding to the same environmental indicator of different tree areas, the distance between different tree areas is adjusted to obtain all sub-tree clusters, and the initially obtained tree clusters are further finely divided; finally, in each sub-tree cluster, according to the environmental information under the environmental indicator corresponding to the maximum difference separation weight of all tree areas in the tree cluster to which it belongs, each tree area is image enhanced and labeled for classification. The present invention first preliminarily clusters the tree areas based on the morphological characteristics of the trees, further analyzes the differentiated performance of the environments of different trees in the same cluster, determines the ability of each environmental indicator to distinguish tree species, and then finely classifies the trees in the cluster again, and enhances the tree areas in different sub-clusters to different degrees, thereby improving the collection effect of land cover remote sensing information. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 A flowchart of a method for collecting land cover remote sensing information for complex natural scenes provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0028] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of a method for collecting land cover remote sensing information for complex natural scenes proposed in accordance with the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0029] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0030] The following describes in detail a method for collecting land cover remote sensing information for complex natural scenes provided by the present invention with reference to the accompanying drawings.

[0031] See also Figure 1 , which shows a method flow chart of a method for collecting land cover remote sensing information for complex natural scenes provided by one embodiment of the present invention, specifically comprising:

[0032] Step S1: Acquire all vegetation areas in the remote sensing image of the area to be measured, and obtain the environmental information of each vegetation area under each environmental indicator, as well as all tree areas therein.

[0033] It should be noted that the implementation scenario targeted by the embodiment of the present invention is the collection of remote sensing information of land cover in summer.

[0034] In one embodiment of the present invention, a region to be surveyed is first determined, and high-resolution original remote sensing images of the region are acquired using the Natural Resources Satellite Remote Sensing Cloud Service Platform. The original remote sensing images are then preprocessed using the improved cloud mask algorithm from Collection 6, including reprojection, resampling, smoothing, filtering, and null value filtering, to obtain high-quality remote sensing images for subsequent analysis. It should be noted that the acquisition and preprocessing of remote sensing images are conventional techniques and will not be further elaborated.

[0035] Since the embodiment of the present invention collects remote sensing information of land cover in complex natural scenes, it mainly focuses on the distribution of land vegetation and the classification and identification of vegetation. However, in addition to vegetation, the area to be measured may also include land, water bodies, buildings, etc.; therefore, after obtaining the remote sensing image of the area to be measured, the embodiment of the present invention further obtains all vegetation areas in the remote sensing image.

[0036] In one embodiment of the present invention, considering that vegetation in remote sensing images in summer has a more special green appearance than other covering objects; since the acquired remote sensing images are RGB images, the channel value of each pixel point in the G channel is obtained in the remote sensing image, and then the pixel points with G channel values ​​greater than a preset threshold such as 180 are taken as target pixel points, and the target pixel points are subjected to regional connectivity detection, and each connected domain is regarded as a vegetation area.

[0037] It should be noted that, in other embodiments, the implementer may also obtain vegetation areas based on technical means such as threshold segmentation or manual labeling segmentation, which, along with obtaining channel values ​​and regional connectivity detection, are existing technical means and will not be described in detail.

[0038] After obtaining the vegetation area, we can further obtain all the tree areas within it, where the tree area refers to the crown area of ​​a single vegetation, in preparation for subsequent classification and labeling of tree species types.

[0039] Preferably, in one embodiment of the present invention, considering that there are gaps between trees and there will be obvious color changes between the gaps and the leaves, all edges can be obtained based on edge detection, and further closed edges can be obtained. The closed edges may be areas corresponding to the tree crowns or areas corresponding to the gaps. Furthermore, considering that the area corresponding to the tree crowns is larger than the area of ​​the gaps, the method for obtaining the tree areas includes:

[0040] Perform edge detection on each vegetation area to obtain all closed edges, and take the area enclosed by each closed edge as a suspected tree area; take the average area of ​​all suspected tree areas as the reference area, and take the suspected tree areas smaller than the reference area as tree areas.

[0041] As an example, taking any vegetation area as an example, first obtain all edges based on the Canny operator, then perform closure detection on each edge to obtain all closed edges, and then obtain all suspected tree areas; then take the total number of pixels in the area as the area, and then obtain the reference area, so as to filter out the tree area.

[0042] It should be noted that, in other embodiments, the implementer may also use a watershed algorithm or a deep learning-based segmentation algorithm to obtain all tree areas in each vegetation area. This, along with Canny edge detection and closure detection, are already existing technical means and will not be described in detail.

[0043] Taking into account that different geographical environments may lead to inconsistent tree growth within vegetation areas, in one embodiment of the present invention, environmental information of each vegetation area under each environmental indicator is further obtained to characterize the geographical environment of each vegetation area, in preparation for the subsequent division of tree areas based on the different performances of trees under different environmental indicators.

[0044] In a preferred embodiment of the present invention, the environmental indicators include at least regional altitude, forest density, and distance to water source.

[0045] It should be noted that the altitude can be obtained based on open source DEM data, and the forest density can be obtained by calculating the NDVI value based on remote sensing images. Then, the water bodies around the measured area can be obtained based on the open source water resource management platform to estimate the distance to the water source. The acquisition process is all existing technology and will not be repeated here. Implementers can also customize the types and quantities of environmental indicators, such as adding tree height, soil pH, etc.

[0046] Step S2: In the remote sensing image, all tree areas are clustered according to the morphological and color characteristics of each tree area to obtain all tree clusters; within the preset neighborhood of each tree area, under each environmental indicator, the surrounding environment complexity of each tree area is obtained according to the difference in environmental information of all vegetation areas.

[0047] Taking into account that remote sensing images are images from a bird's-eye view, they mainly show the crowns of trees, and the crowns of different types of trees have different shapes, which is first reflected in the different sizes of the crowns. For example, the crowns of trees are larger than those of shrubs. Secondly, it is reflected in the clustering of the crowns. For example, the branches of pine trees extend outward and the extension degree of different branches is different. Its crown is more layered when viewed from a bird's-eye view, while the branches of apple trees are more clustered, and it appears more rounded and smooth when viewed from a bird's-eye view. Taking into account that although different trees all have the characteristic of green, there are still large differences in the color of the leaves of different trees. Therefore, the embodiment of the present invention first preliminarily clusters all tree areas in the remote sensing image to obtain all tree clusters. The tree areas in each tree cluster are tree areas suspected to be of the same species.

[0048] Preferably, in one embodiment of the present invention, considering that the a-axis in the CIELAB color space can represent the green characteristics of the tree area, it can help to more clearly observe the color differences between different tree areas; the area of ​​the tree region can reflect the size of the crown, and the corner point is a characteristic point of a significant turning point or change point, which is usually located at the edge of an object or where the shape changes. Therefore, the number of corner points in a clustered and rounded crown area should be smaller than the number of corner points in a relatively layered crown area. Therefore, the crown shape can be evaluated based on the number of corner points in the tree area, and the crown performance of each tree area can be characterized by combining color characteristics and area; and considering that clustering can cluster similar trees into a cluster, based on this, the method for obtaining tree clusters includes:

[0049] The remote sensing image was converted into CIE color space, and the mean a-axis value of all pixels in each tree area was used as the greenness value; the number of corner points in each tree area in the remote sensing image was used as the crown shape parameter;

[0050] The triplet constructed by the greenness value, crown shape parameter and area size of each tree area is used as the characteristic parameter of each tree area. Based on the distance clustering algorithm, the characteristic parameters corresponding to all tree areas in the remote sensing image are clustered, and the corresponding tree clusters are obtained according to the clustering of the characteristic parameters.

[0051] As an example, we first obtain the greenness value of each tree area, then perform Harris corner detection on each tree area to obtain all corner points, thereby obtaining the crown shape parameters of each tree area; then construct triples to obtain the characteristic parameters of each tree area. Considering that the ISODATA clustering algorithm does not require a pre-set number of clusters, the characteristic parameters corresponding to all tree areas in the remote sensing image are clustered based on the ISODATA clustering algorithm to obtain several clusters, where all the tree areas corresponding to the characteristic parameters in each cluster constitute a tree cluster;

[0052] To facilitate understanding, implementers can also directly construct a spatial coordinate system and map the triplet corresponding to each tree area into the spatial coordinate system, thereby clustering the tree areas based on the ISODATA clustering algorithm.

[0053] It should be noted that in other examples, implementers can also use the mode instead of the mean to obtain the greenness value; considering that the crown size of different tree areas may affect the number of corner points, after obtaining the number of corner points of each tree area, it can be further divided by the area of ​​the corresponding tree area to obtain the number of corner points per unit area, eliminating the influence of the crown size and accurately obtaining the crown shape parameters; implementers can also use other clustering algorithms such as the K-means clustering algorithm, which, together with CIELAB-based a-axis value for each pixel, Harris corner detection and ISODATA clustering, are all existing technologies and will not be repeated here.

[0054] Considering that the area to be tested may have a variety of complex terrains or complex natural scenes, such as the transition zone between mountains and hills, and different species of trees adapt to different growth environments, resulting in differences in the types and growth conditions of trees in different vegetation areas; considering that the environmental information of each vegetation area under each environmental indicator has been obtained, the surrounding environmental complexity of each tree area can be further obtained based on the differences in the environmental information of all vegetation areas under each environmental indicator within the preset neighborhood of each tree area;

[0055] The complexity of the surrounding environment provides a certain reference for the growth geographical environment performance of each tree area under each environmental indicator, and prepares for the subsequent evaluation of the impact of each environmental indicator on tree growth under each environmental indicator, combined with the distribution of tree areas in the same cluster under similar surrounding environmental complexity, to obtain the difference separation weight.

[0056] Preferably, in one embodiment of the present invention, the preset neighborhood of each tree area is first determined, and the preset neighborhood of each tree area is all the adjacent vegetation areas of the vegetation area to which each tree area belongs, so as to subsequently analyze the complexity of the surrounding environment of each tree area.

[0057] Preferably, in one embodiment of the present invention, considering that the standard deviation can be used to understand the distribution characteristics of the data, under each environmental indicator, if the standard deviation of the environmental information corresponding to all adjacent vegetation areas is larger, it means that the degree of dispersion of the environmental information is greater and the environmental information is more complex; therefore, the method for obtaining the complexity of the surrounding environment includes:

[0058] Within the preset neighborhood of each tree area, the standard deviation of the environmental information of all adjacent vegetation areas under each environmental indicator is used as the surrounding environmental complexity of each tree area under the corresponding environmental indicator. It should be noted that obtaining the standard deviation is a well-known technique, and implementers can also use discrete parameters such as variance to measure surrounding environmental complexity, which will not be explained here.

[0059] In other embodiments, the implementer may also expand the scope of analysis, for example, by splicing remote sensing images of areas surrounding the area to be measured, such as within an 8-neighborhood, with remote sensing images of the area to be measured, and obtaining all vegetation areas and corresponding environmental information, thereby analyzing and obtaining the complexity of the surrounding environment of each tree area; splicing remote sensing images is already an existing technical means and will not be repeated.

[0060] Step S3: In each tree cluster, the difference separation weight of each environmental indicator is obtained according to the difference in the number of tree areas corresponding to each surrounding environment complexity, and the distance between different tree areas is adjusted according to the difference in the environmental information corresponding to the same environmental indicator of different tree areas to obtain all sub-tree clusters; in each sub-tree cluster, the image of each tree area is enhanced and labeled according to the environmental information under the environmental indicator corresponding to the maximum difference separation weight of all tree areas in the tree cluster to which it belongs.

[0061] Taking into account that during the initial cluster classification, trees of different species but with similar crown performance may be divided into the same tree cluster; and considering that the growth environment of trees of the same species is relatively similar, in each tree cluster, the distribution of the surrounding environment complexity of the tree area under each environmental indicator can help reflect the differences in the geographical environment or growth environment corresponding to trees with similar crowns; that is, under different surrounding environment complexities under a certain environmental indicator, the more obvious the difference in the number of tree areas in the same tree cluster, the more likely the tree areas in the cluster are concentrated in a certain geographical environment, and the stronger the ability of the environmental indicator to distinguish tree species with similar growth environments, the easier it is to separate different tree species;

[0062] Therefore, in each tree cluster, the embodiment of the present invention obtains the difference separation weight of each environmental indicator according to the difference in the number of tree areas corresponding to each surrounding environment complexity; the difference separation weight reflects the ability of each environmental indicator to help distinguish different tree species in each tree cluster.

[0063] Preferably, in one embodiment of the present invention, considering that in each tree cluster, under each environmental indicator, the surrounding environment complexity is used as the horizontal axis parameter and the number of tree areas corresponding to the surrounding environment complexity is used as the vertical axis parameter, a characteristic curve is fitted. By analyzing the degree of fluctuation of the characteristic curve, it is possible to help analyze the distribution of the number of tree areas corresponding to each type of surrounding environment complexity under each environmental indicator, and then evaluate the difference separation weight. Based on this, the method for obtaining the difference separation weight includes:

[0064] In each tree cluster, under each environmental indicator, a characteristic curve was fitted with the surrounding environment complexity as the horizontal axis parameter and the number of tree areas corresponding to the surrounding environment complexity as the vertical axis parameter; based on the degree of fluctuation of the characteristic curve, the difference separation weight under the corresponding environmental indicator was obtained.

[0065] In a preferred embodiment of the present invention, considering that the more extreme value points in the characteristic curve and the greater the rate of change between adjacent extreme value points, the greater the fluctuation of the characteristic curve, the method for obtaining the difference separation weight under the corresponding environmental indicator according to the degree of fluctuation of the characteristic curve includes:

[0066] In the characteristic curve, the difference separation weight under the corresponding environmental index is obtained according to the change rate between adjacent extreme points and the total number of extreme points.

[0067] As an example, in the characteristic curve, all extreme points are first obtained, and then the vertical axis parameter difference between adjacent extreme points, that is, the absolute value of the difference between the number of corresponding tree areas, is divided by the horizontal axis parameter difference between adjacent extreme points, that is, the absolute value of the difference between the corresponding surrounding environment complexity, to obtain the change rate between the corresponding adjacent extreme points; then all the change rates are summed, the sum is multiplied and combined with the total number of extreme points, and the product is mapped to the sigmoid function for normalization to obtain the difference separation weight under the corresponding environmental indicator of the characteristic curve.

[0068] It should be noted that fitting characteristic curves, obtaining extreme points and normalization are all existing technical means and will not be described in detail.

[0069] In each tree cluster, after obtaining the difference separation weight of each environmental indicator, we can further combine the similar characteristics of the living geographical environment of trees of the same species to analyze the differences in environmental information corresponding to each environmental indicator in different tree areas, and adjust the distance between different tree areas in combination with the difference separation weight, so as to perform secondary clustering within the tree cluster and obtain all sub-tree clusters within it, where the tree areas in the sub-tree clusters are the tree areas of the same tree species.

[0070] Preferably, in one embodiment of the present invention, considering that in each tree cluster, the characteristics of the cluster center can reflect the average characteristics of all tree areas, the greater the difference in environmental information corresponding to each tree area relative to the cluster center under each environmental indicator, the greater the difference, indicating that the two may not be in similar growth environments. At the same time, if the ability of the environmental indicator to distinguish tree species, that is, the difference separation weight, is greater, it means that the environmental information difference provided under the environmental indicator has a greater reference value for distinguishing tree species in different growth environments. Based on this, the distance between each tree area and the cluster center can be adjusted to increase the difference between tree areas in different living environments, so as to accurately distinguish or classify them. Therefore, the method for obtaining sub-tree clusters includes:

[0071] In each tree cluster, under each environmental indicator, the difference separation weight is used to weight the difference in environmental information between each tree area and the cluster center. The corrected distance between each tree area and the cluster center is obtained based on the weighted sum value under all environmental indicators; all tree areas in each tree cluster are clustered again according to the corrected distance to obtain all sub-tree clusters.

[0072] As an example, in each tree cluster, under each environmental indicator, the difference in environmental information corresponding to each tree area and the cluster center is measured in the form of the square of the absolute value of the difference, and then the difference separation weight is multiplied by the square of the absolute value of the difference of the corresponding environmental information to obtain the weighted sum value under each environmental indicator, and the weighted sum value is squared to obtain the corrected distance between each tree area and the cluster center; then, according to the corrected distance of all tree areas in the tree cluster relative to the cluster center, all tree areas are mapped to the spatial model, and then the tree areas in the tree cluster are secondary clustered based on the ISODATA clustering algorithm to obtain all sub-tree clusters.

[0073] It should be noted that spatial mapping and clustering based on the corrected distance is an existing technology and will not be described in detail. Before using the difference separation weight for weighted summation, the difference separation weight under each environmental indicator needs to be normalized, and the sum of the difference separation weights under all environmental indicators is 1. Assuming that there are three environmental indicators abc, taking environmental indicator a as an example, the normalization method of its difference separation weight is , where abc are the serial numbers of environmental indicators, is the difference separation weight of the a-th environmental indicator, is the difference separation weight of the b-th environmental indicator, is the difference separation weight of the cth environmental indicator.

[0074] In another embodiment of the present invention, the implementer may also replace the cluster center with other tree areas in each tree cluster, that is, under each environmental indicator, use the difference separation weight to weight the difference between the corresponding environmental information of two different tree areas, and then use the weighted sum value under all environmental indicators as the corrected distance between the two different tree areas; obtain the corrected distance between all two different tree areas, and then map it to the cluster in the spatial model to obtain the sub-tree area, which is the same as the above-mentioned acquisition method and will not be repeated.

[0075] After obtaining all sub-tree clusters, the tree areas of different tree species corresponding to different sub-tree clusters can be further enhanced to different degrees to enhance the distinction between tree areas of different tree species; considering that in each tree cluster, the maximum difference separation weight corresponds to the environmental information under the environmental index, and the ability to distinguish sub-tree areas of different tree species is the strongest, the embodiment of the present invention will enhance each tree area according to the environmental information under the environmental index corresponding to the maximum difference separation weight of all tree areas in the tree cluster to which it belongs.

[0076] Preferably, in one embodiment of the present invention, the method for enhancing each tree region includes:

[0077] In each sub-tree cluster, under the environmental index corresponding to the maximum difference separation weight in the tree cluster to which it belongs, the mean of the environmental information of the vegetation area to which all tree areas belong is normalized, and the normalized result is used as the enhancement coefficient corresponding to all tree areas in the corresponding sub-tree cluster; the corresponding tree area is enhanced using the enhancement coefficient.

[0078] As an example, the mean is mapped to a sigmoid function for normalization. The implementer may also use other normalization functions to obtain the enhancement coefficients of all tree areas in each tree planting area, and then use the enhancement coefficients to enhance the brightness or color of the corresponding trees. This is already an existing technology and will not be described in detail.

[0079] In another embodiment of the present invention, the implementer may also directly regard each sub-tree cluster as a collection of tree areas under a tree species, and then mark the tree areas in different sub-tree clusters with different colors for classification, which will not be repeated here.

[0080] In summary, the present invention first preliminarily obtains all tree clusters based on the morphological and color characteristics of each tree area; further, under each environmental indicator, the surrounding environment complexity of each tree area is analyzed and obtained; then, in each tree cluster, the difference separation weight of each environmental indicator is obtained based on the difference in the number of tree areas corresponding to each surrounding environment complexity, and the distance between different tree areas is adjusted based on the difference in environmental information corresponding to the same environmental indicator, to obtain all sub-tree clusters; finally, each tree area in each sub-tree cluster is image enhanced and labeled for classification. The present invention first preliminarily clusters tree areas based on the morphological characteristics of the trees, further analyzes the differentiated performance of the environments in which different trees in the same cluster are located, determines the ability of each environmental indicator to distinguish tree species, and then further fine-classifies the trees in the cluster and enhances the tree areas in different sub-clusters to varying degrees, thereby improving the collection effect of land cover remote sensing information.

[0081] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0082] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A land cover remote sensing information collection method for complex natural scenes, characterized by: The method comprises: Obtain all vegetation areas in the remote sensing image of the area to be measured, and obtain environmental information of each vegetation area under each environmental indicator, as well as all tree areas within the vegetation area; the environmental indicators include at least regional altitude, tree density, and distance to water source; In remote sensing images, all tree areas are clustered according to their morphological and color characteristics to obtain all tree clusters. Within the preset neighborhood of each tree area, under each environmental indicator, the surrounding environmental complexity of each tree area is obtained based on the differences in the environmental information of all vegetation areas. In each of the tree clusters, based on the difference in the number of tree areas corresponding to each of the surrounding environment complexities, a difference separation weight for each environmental indicator is obtained, and combined with the difference between the environmental information corresponding to the same environmental indicator of different tree areas, the distances between different tree areas are adjusted to obtain all sub-tree clusters; in each sub-tree cluster, based on the environmental information corresponding to the environmental indicator with the maximum difference separation weight of all tree areas in the tree cluster, each tree area is image enhanced and labeled for classification; The method for obtaining the subtree clusters includes: In each of the tree clusters, under each environmental indicator, the difference separation weight is used to weight the difference in environmental information between each tree region and the cluster center, and the weighted sum value under all environmental indicators is obtained to obtain a corrected distance between each tree region and the cluster center; all tree regions in each of the tree clusters are secondary clustered according to the corrected distance to obtain all sub-tree clusters; The methods used to enhance each tree region include: In each sub-tree cluster, the mean value of the environmental information under the environmental index corresponding to the maximum difference separation weight in the tree cluster to which it belongs is normalized, and the normalized result is used as the enhancement coefficient corresponding to all tree areas in the corresponding sub-tree cluster; the corresponding tree area is enhanced using the enhancement coefficient.

2. The method for collecting land cover remote sensing information for complex natural scenes according to claim 1, characterized in that: The method for obtaining the tree clusters includes: The remote sensing image was converted into CIE color space, and the mean a-axis value of all pixels in each tree area was used as the greenness value; the number of corner points in each tree area in the remote sensing image was used as the crown shape parameter; The triplet constructed by the greenness value, crown shape parameter and area size of each tree area is used as the characteristic parameter of each tree area, and based on the distance clustering algorithm, the characteristic parameters corresponding to all tree areas in the remote sensing image are clustered, and the corresponding tree clusters are obtained according to the clustering of the characteristic parameters.

3. The method for collecting land cover remote sensing information for complex natural scenes according to claim 1, characterized in that: The preset neighborhood of each tree area is all the bordering vegetation areas of the vegetation area to which each tree area belongs.

4. The method for collecting land cover remote sensing information for complex natural scenes according to claim 3, characterized in that: The method for obtaining the complexity of the surrounding environment includes: In a preset neighborhood of each tree area, the standard deviation of the environmental information of all adjacent vegetation areas under each environmental indicator is used as the surrounding environment complexity of each tree area under the corresponding environmental indicator.

5. The method for collecting land cover remote sensing information for complex natural scenes according to claim 1, characterized in that: The method for obtaining the difference separation weight includes: In each of the tree clusters, under each environmental indicator, a characteristic curve is fitted with the complexity of the surrounding environment as the horizontal axis parameter and the number of tree areas corresponding to the complexity of the surrounding environment as the vertical axis parameter; and based on the degree of fluctuation of the characteristic curve, the difference separation weight under the corresponding environmental indicator is obtained.

6. The method for collecting land cover remote sensing information for complex natural scenes according to claim 5, characterized in that: According to the fluctuation degree of the characteristic curve, the method for obtaining the difference separation weight under the corresponding environmental indicator includes: In the characteristic curve, the difference separation weight under the corresponding environmental index is obtained according to the change rate between adjacent extreme value points and the total number of extreme value points.

7. The method for collecting land cover remote sensing information for complex natural scenes according to claim 1, characterized in that: The method for obtaining the tree area includes: Edge detection is performed on each vegetation area to obtain all closed edges, and the area enclosed by each closed edge is regarded as a suspected tree area; the average area of ​​all suspected tree areas is used as a reference area, and the suspected tree areas smaller than the reference area are regarded as tree areas.

Images