Forest combustible carrying capacity measuring and calculating method and system

Abstract

The invention discloses a forest combustible load measuring and calculating method and system, and the method comprises the steps: carrying out the segmentation processing of an image corresponding to each region of a forest according to the RGB ranges of the forest combustible in different layers, and obtaining the images corresponding to different layers of each region of the forest; calculating parameter information of different layers of each region of the forest based on the images corresponding to the different layers of each region of the forest; biomass models of different levels are adopted to calculate biomass information of different levels of each region of the forest based on the parameter information sets of different levels of each region of the forest; based on the biomass information of the different layers of each region of the forest, obtaining the combustible loading capacity of the different layers of each region of the forest; and calculating the combustible loading capacity of each region of the forest based on the combustible loading capacity of different layers of each region of the forest. By utilizing the scheme of the invention, the influence of environmental factors can be avoided, the influence of the tree canopy layer on the data acquisition of other levels is also avoided, the required time is short, and the workload is small.

Description

Technical Field

[0001] The present application generally relates to the technical field of fuel load estimation, and more specifically, to a method and system for estimating forest fuel load. Background Art

[0002] The forest combustible load refers to the absolute dry weight of all forest combustibles per unit area. The forest combustible load has a significant impact on the risk prevention and control of forest fires. When the combustible load in the forest reaches a high level, serious forest fires may occur. Once excessive combustible accumulation catches fire, the fire will be extremely fierce and difficult to control. In addition, a high load of combustibles may also lead to crown fires. This type of fire is extremely destructive and can cause devastating damage to forest ecosystems. Therefore, it is necessary to effectively measure the forest combustible load so that the combustibles can be cleaned up in a timely manner based on the measurement results to reduce the risk of fire.

[0003] In the prior art, forest fuel load is mainly measured by plot survey method and remote sensing estimation method. The plot survey method is to collect all forest fuels in the selected forest plot, and then take them to the laboratory to dry them to be manually weighed and analyzed or calculated using mathematical models. However, the workload of the plot survey method is huge, especially in large forests, collecting, drying and weighing all fuels is very time-consuming and laborious. In addition, during the collection process, omissions or repeated collections are prone to occur, which affects the accuracy of the results. At the same time, the spatial representativeness is limited, and the results of one plot are difficult to represent the situation of the entire forest area, because the distribution of forest fuels is very heterogeneous in space. Even if multiple plots are set up, it is difficult to fully cover all changes. The remote sensing estimation method uses satellites or aerial optical sensors to obtain spectral images of the forest, and estimates the fuel load based on the differences in the reflection characteristics of different vegetation types and fuels in optical bands (such as visible light and near-infrared bands). However, the data obtained by the remote sensing estimation method is not accurate, and data acquisition is easily affected by weather, time, etc.

[0004] In view of this, there is an urgent need to provide a forest fuel load calculation solution so that the data will not be affected by environmental factors during the forest fuel load calculation process, and it takes a short time, has a small workload, and is simple and convenient to operate. Summary of the invention

[0005] In order to at least solve one or more of the technical problems mentioned above, the present application proposes a forest fuel load calculation scheme in multiple aspects.

[0006] In the first aspect, the present application provides a method for measuring forest combustible load, including: segmenting images corresponding to each forest area according to the RGB range of forest combustibles at different levels to obtain images corresponding to different levels of each forest area; calculating parameter information of different levels of each forest area based on the images corresponding to different levels of each forest area; using different levels of biomass models to calculate biomass information of different levels of each forest area based on parameter information sets of different levels of each forest area; obtaining combustible loads of different levels of each forest area based on biomass information of different levels of each forest area; calculating the combustible loads of each forest area based on the combustible loads of different levels of each forest area.

[0007] In some embodiments, the different layers include a tree layer, a shrub layer, a herb layer, and a surface litter and humus layer.

[0008] In some embodiments, parameter information at different levels includes: the average height of the tree layer, the average breast diameter of the tree layer, the number of trees in the tree layer, the density of the tree layer, the average height of the shrub layer, the average base diameter of the shrub layer, the density of the shrub layer, the average height of the herb layer and the density of the herb layer.

[0009] In some embodiments, in the process of calculating the average height of the tree layer, the average height of the shrub layer, and the average height of the herb layer, the following steps are performed: grayscale processing and image binarization processing are performed on the images corresponding to the layers whose average heights need to be calculated in sequence; all contours in the images after grayscale processing and binarization processing are searched; the reference object contours in all contours are matched using a preset index; the pixel height of the reference object contour is obtained; and the average height corresponding to the corresponding layer is obtained based on the pixel height of the reference object contour.

[0010] In some embodiments, obtaining the average height corresponding to the corresponding level based on the pixel height of the reference object contour includes: traversing all contours except the reference object contour to obtain the pixel heights of the contours except the reference object contour; calculating the ratio between the pixel height of each other contour and the pixel height of the reference object contour; obtaining the actual height of each other contour by comparing the actual height of the reference object with the ratio; and obtaining the average height corresponding to the corresponding level by calculating the average of the actual heights of each other contour.

[0011] In some embodiments, in the process of calculating the average breast diameter of the tree layer, the following steps are performed: based on the image corresponding to the tree layer, all contours in the image are acquired, wherein all contours include reference object contours and multiple non-reference object contours; the ratio between the pixel width of the reference object contour and the pixel width of each non-reference object contour is acquired; the actual breast diameter of each non-reference object contour is obtained by comparing the actual breast diameter of the reference object with the ratio; and the average breast diameter of the tree layer is obtained by calculating the average of the actual breast diameters of each non-reference object contour.

[0012] In some embodiments, in the process of calculating the canopy density of the tree layer, the canopy density of the shrub layer, and the canopy density of the herb layer, the following steps are performed: the images corresponding to the layers for which the canopy density needs to be calculated are subjected to grayscale processing, contrast enhancement processing, and image binarization processing in turn; the ratio of the number of white pixels in the image after grayscale processing, contrast enhancement processing, and image binarization processing to the total number of pixels is calculated to obtain the canopy density of the corresponding layer.

[0013] In some embodiments, parameter information at different levels also includes moisture content of combustibles at different levels. In the process of calculating the moisture content of combustibles at different levels in various forest areas, the following steps are performed: images corresponding to different levels in various forest areas are converted to HSV color space; areas in the HSV color space that meet the set HSV threshold range are extracted; the ratio of pixels in the extracted area to the total pixels of the corresponding image is calculated, and the moisture content of the combustibles at the corresponding level is obtained based on the ratio.

[0014] In some embodiments, the biomass information at different levels includes: tree trunk biomass, tree branch biomass, tree leaf biomass, shrub branch biomass, shrub leaf biomass, herbaceous layer biomass, surface litter and humus layer biomass; the first calculation formula is used to obtain the combustible load of the tree layer in each area of ​​the forest based on the tree trunk biomass, tree branch biomass and tree leaf biomass, and the first calculation formula is: M1=[W 乔干 +W 乔枝 +W 乔叶 ]×n1×(1-P1), where W 乔干 is the trunk biomass of the tree layer, W 乔干 =2.52+0.12(D 2 ×h1), W 乔枝 is the tree layer branch biomass, W 乔枝 =0.06(D 2 ×h1) 0.73 , W 乔叶 is the leaf biomass of tree layer, W 乔叶 =17.459×h1 0.482, h1 is the average height of the tree layer, D is the average breast diameter of the tree layer, n1 is the number of trees in the tree layer, and P1 is the moisture content of the tree layer;

[0015] The second calculation formula is used to obtain the fuel load of the shrub layer in each area of ​​the forest based on the shrub layer branch biomass and shrub layer leaf biomass. The second calculation formula is: M2 = [W 灌枝 +W 灌叶 ]×S 灌 ×U 灌 ×(1-P2), where W 灌枝 is the shrub layer branch biomass, W 灌枝 =0.28(B 2 ×h2) 0.74 , W 灌叶 is the leaf biomass of the shrub layer, W 灌叶 =10.463×h2 0.476 , h2 is the average height of the shrub layer, B is the average base diameter of the shrub layer, S 灌 is the area of ​​the shrub layer, U 灌 is the density of the shrub layer, P2 is the moisture content of the shrub layer; the third calculation formula is used to obtain the combustible load of the herbaceous layer in each area of ​​the forest based on the biomass of the herbaceous layer. The third calculation formula is: M3 = W 草 ×(1-P3),W 草 is the herbaceous layer biomass, W 草 =0.40×[(S 草 ×U 草 ) 2 ×h3] 0.56 , S 草 is the area of ​​the herbaceous layer, U 草 is the canopy density of the herbaceous layer, h3 is the average height of the herbaceous layer, and P3 is the moisture content of the herbaceous layer; the fourth calculation formula is used to obtain the combustible load of the surface litter and humus layer in each area of ​​the forest based on the biomass of the surface litter and humus layer. The fourth calculation formula is: M4=W 总 ×S 总 ×P4,W 总 is the biomass of surface litter and humus layer, W 总 =h 腐 +h 地 ×t,h 腐 is the thickness of humus, h 地 is the thickness of the litter on the ground, t is the number of years, S 总 is the area of ​​the surface litter and humus layer, and P4 is the moisture content of the surface litter and humus layer.

[0016] In the second aspect, the present application provides a forest combustible load measurement system, which uses the forest combustible load measurement method as described in any embodiment of the first aspect to perform forest combustible load measurement, and the system includes: an image processing module, which is used to segment the images corresponding to each forest area according to the RGB range of forest combustibles at different levels, and obtain images corresponding to different levels of each forest area; a parameter acquisition module, which is used to calculate parameter information of different levels of each forest area based on the images corresponding to different levels of each forest area; a biomass calculation module, which is used to calculate biomass information of different levels of each forest area based on parameter information sets of different levels of each forest area using biomass models of different levels; a combustible load calculation module, which is used to obtain the combustible load of different levels of each forest area based on the biomass information of different levels of each forest area, and calculate the combustible load of each forest area based on the combustible load of different levels of each forest area.

[0017] Through the forest combustible load calculation scheme provided above, the embodiment of the present application obtains images corresponding to different layers of each forest area by segmenting the images corresponding to each area of ​​the forest. The weight of combustibles in each layer can be calculated according to the characteristics of different plant layers, and the weight of forest combustibles in a certain area can be obtained by cumulative averaging. It will not be affected by environmental factors, nor will it be affected by the canopy layer on the acquisition of data at other levels. It also requires a short time and a small workload. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easy to understand. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0019] Figure 1 An exemplary flow chart of a method for calculating forest fuel load according to an embodiment of the present application is shown;

[0020] Figure 2 An exemplary flow chart of calculating the average height of the tree layer, shrub layer and herb layer according to an embodiment of the present application is shown;

[0021] Figure 3 An exemplary flow chart of obtaining the average height corresponding to the corresponding layer based on the pixel height of the reference object contour according to an embodiment of the present application is shown;

[0022] Figure 4 An exemplary flow chart of calculating the average breast height diameter of a tree layer according to an embodiment of the present application is shown;

[0023] Figure 5An exemplary flow chart of calculating canopy closure according to an embodiment of the present application is shown;

[0024] Figure 6 An exemplary flow chart of calculating the moisture content of an embodiment of the present application is shown;

[0025] Fig. 7A An image of a forest area according to an embodiment of the present application is shown;

[0026] Figure 7B An embodiment of the present application is shown by dividing Fig. 7A The image corresponding to the tree layer is obtained from the image;

[0027] Figure 7C An embodiment of the present application is shown by dividing Fig. 7A The image corresponding to the shrub layer is obtained from the image;

[0028] Fig.7D An embodiment of the present application is shown by dividing Fig. 7A The image corresponding to the herb layer is obtained from the image;

[0029] Fig. 7E An embodiment of the present application is shown by dividing Fig. 7A The image corresponding to the surface litter and humus layer is obtained from the image;

[0030] Figure 8 An exemplary structural block diagram of a forest fuel load estimation system according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0032] It should be understood that the terms "include" and "comprising" used in the specification and claims of the present application indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0033] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this application specification and claims, unless the context clearly indicates otherwise, the singular forms of "a", "an" and "the" are intended to include plural forms. It should also be further understood that the term "and / or" used in this application specification and claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0034] As used in this specification and claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0035] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0036] Figure 1 An exemplary flow chart of a forest fuel load estimation method 100 according to an embodiment of the present application is shown.

[0037] like Figure 1 As shown, in step S110, the images corresponding to the various forest areas are segmented according to the RGB ranges of the forest combustibles at different levels to obtain the images corresponding to the various levels of the forest areas.

[0038] In the embodiment of the present application, there are multiple images corresponding to each area of ​​the forest, and the multiple images are taken at multiple different angles of the corresponding area of ​​the forest. Specifically, the number of images corresponding to each area of ​​the forest and the shooting angles can be set as needed, and the present application does not limit this.

[0039] After processing a plurality of images corresponding to the corresponding forest area, the average value of the combustible load corresponding to each image is calculated to obtain the combustible load corresponding to the forest area.

[0040] By taking multiple images at multiple different angles in the corresponding area of ​​the forest and calculating the average value of the combustible load corresponding to each image, the combustible load corresponding to the forest area can be obtained. This can more comprehensively reflect the combustible load situation of the forest area, and by calculating the average value, the error can be further reduced, thereby improving the accuracy of the combustible load estimation.

[0041] In an embodiment of the present application, before the images corresponding to each area of ​​the forest are segmented, the images corresponding to each area of ​​the forest are subjected to feature extraction. Specifically, in the feature extraction process, first, the deeplabv3 model preprocesses the images corresponding to each area of ​​the forest, and the preprocessing specifically includes adjusting the size and normalizing the pixel value so that the images corresponding to each area of ​​the forest meet the input requirements of the deeplabv3 model. Subsequently, the preliminary convolution layer of the deeplabv3 model extracts the low-level features including color and edge corresponding to the images corresponding to each area of ​​the forest, and then extracts the multi-scale features corresponding to the images corresponding to each area of ​​the forest based on different sampling rates through the hole convolution in the spatial pyramid pooling module, capturing from the macroscopic overall structure of the forest to the microscopic local details. At the same time, the extracted low-level features and scale features are fused through multi-layer convolution, so that the low-level features are gradually converted into high-level semantic features containing rich semantic information, so that the model can understand the semantic categories of different objects in the image.

[0042] In the embodiments of the present application, the different layers include a tree layer, a shrub layer, a herb layer, and a surface litter and humus layer.

[0043] In the embodiment of the present application, during the segmentation process of the images corresponding to each forest area, the RGB range of each pixel is judged by a classification algorithm based on the extracted and fused feature information, and finally the segmented image is output.

[0044] Specifically, the RGB range of the tree layer is: the range of R (red) channel is 100-255, the range of G (green) channel is 120-255, and the range of B (blue) channel is 80-200. The RGB range of the shrub layer is: the range of R (red) channel is 800-200, the range of G (green) channel is 100-230, and the range of B (blue) channel is 60-180. The RGB range of the herb layer is: the range of R (red) channel is 60-180, the range of G (green) channel is 80-200, and the range of B (blue) channel is 40-160. The RGB range of the surface litter and humus layer is: the range of R (red) channel is 30-140, the range of G (green) channel is 30-160, and the range of B (blue) channel is 20-120.

[0045] Specifically, all image pixels whose RGB range is within the RGB range of the tree layer are divided into the segmented tree layer image, all image pixels whose RGB range is within the RGB range of the shrub layer are divided into the segmented shrub layer image, all image pixels whose RGB range is within the RGB range of the herb layer are divided into the segmented herb layer image, and all image pixels whose RGB range is within the RGB range of the surface litter and humus layer are divided into the segmented surface litter and humus layer image. Thus, by segmenting the images corresponding to each area of ​​the forest, the corresponding tree layer image, shrub layer image, herb layer image, and surface litter and humus layer image are obtained.

[0046] In an embodiment of the present application, when redundant parts appear after segmentation processing of the images corresponding to each area of ​​the forest, the redundant parts can be manually eliminated by increasing the contrast. Specifically, in the process of manually eliminating the redundant parts by increasing the contrast, first use a suitable tool or algorithm (such as histogram equalization, adaptive contrast adjustment, etc.) to increase the contrast of the image to help strengthen the distinction between the target object and the background or other non-related elements. Then, the image after contrast adjustment is segmented again. Select a suitable segmentation algorithm according to the specific situation, such as threshold segmentation, edge detection, regional growth method, watershed algorithm, etc. The segmentation results are post-processed to remove small, discontinuous areas by filling holes, smoothing boundaries, etc. Then, repeat the above process until all redundant parts are removed.

[0047] By increasing the contrast, the difference in light and dark between different areas in the image can be enhanced, making the originally difficult-to-identify boundaries clearer. This helps to more accurately identify the boundaries between the various levels of the corresponding forest areas during the image segmentation process and reduce the appearance of redundant parts. By manually removing these redundant parts, the accuracy of image segmentation can be further improved, providing more valuable information for subsequent image analysis and processing.

[0048] After executing step S110, in step S120, parameter information of different levels of each forest area is calculated based on images corresponding to different levels of each forest area.

[0049] In an embodiment of the present application, parameter information at different levels includes: the average height of the tree layer, the average breast diameter of the tree layer, the number of trees in the tree layer, the density of the tree layer, the average height of the shrub layer, the average base diameter of the shrub layer, the density of the shrub layer, the average height of the herb layer and the density of the herb layer, the area of ​​the shrub layer, the area of ​​the herb layer, the thickness of humus, the thickness of surface litter, the area of ​​surface litter and humus layer, etc.

[0050] In the embodiment of the present application, the specific process of calculating the average height of the tree layer, the average height of the shrub layer, and the average height of the herb layer can be found in Figure 2 .

[0051] Figure 2 An exemplary flow chart for calculating the average heights of the tree layer, shrub layer, and herb layer according to an embodiment of the present application is shown.

[0052] like Figure 2 As shown, in step S210, the image corresponding to the level for which the average height needs to be calculated is grayed and binarized in turn. In step S220, all contours in the image after graying and binarization are searched. In step S230, the reference object contours in all contours are matched using a preset index. In step S240, the pixel height of the reference object contour is obtained. In step S250, the average height corresponding to the corresponding level is obtained based on the pixel height of the reference object contour.

[0053] In the embodiment of the present application, during the grayscale processing, an image processing library (such as OpenCV) can be used to convert a color image into a grayscale image.

[0054] In an embodiment of the present application, after the grayscale processing, the grayscale processed image may be subjected to Gaussian blur processing, and then the Gaussian blurred image may be subjected to image binarization processing. Specifically, the noise of the grayscale processed image may be reduced and unnecessary details of the grayscale processed image may be removed by Gaussian blur processing. During the Gaussian blur processing, a convolution operation is performed on the image using a Gaussian distributed weight matrix, thereby achieving a smoothing effect.

[0055] In an embodiment of the present application, during the image binarization process, a threshold is selected to binarize the grayscale image, which sets each pixel in the image to white (foreground) or black (background). Specifically, the aforementioned threshold can be set according to the application scenario to find the best separation effect, and the present application does not limit it here.

[0056] In the embodiments of the present application, existing image processing functions may be used to search for all contours in an image that has been subjected to grayscale processing and binarization processing, and the present application does not impose any limitation thereto.

[0057] In the embodiments of the present application, the aforementioned preset index may be the position of the reference object, the shape of the reference object, the number of the reference object, etc., and the present application does not limit this.

[0058] In the embodiments of the present application, the reference object may be a specially placed standard measuring rod, or it may be a target of a corresponding layer, for example, a tree in the tree layer, a shrub in the shrub layer, an herb in the herb layer, etc., and the present application does not limit this.

[0059] In the embodiment of the present application, the specific process of obtaining the average height corresponding to the corresponding layer based on the pixel height of the reference object contour can be found in Figure 3 .

[0060] Figure 3 An exemplary flow chart of an embodiment of the present application for obtaining an average height corresponding to a corresponding layer based on the pixel height of a reference object contour is shown.

[0061] like Figure 3 As shown, in step S310, all contours except the reference object contour are traversed to obtain the pixel heights of the contours except the reference object contour. In step S320, the ratio between the pixel height of each other contour and the pixel height of the reference object contour is calculated. In step S330, the actual height of each other contour is obtained by comparing the actual height of the reference object with the ratio. In step S340, the average height corresponding to the corresponding level is obtained by calculating the average value of the actual heights of each other contour.

[0062] In the embodiments of the present application, the specific process of calculating the average DBH of the tree layer can be found in Figure 4 .

[0063] Figure 4 An exemplary flow chart for calculating the average breast height diameter of a tree layer according to an embodiment of the present application is shown.

[0064] like Figure 4 As shown, in step S410, all contours in the image are obtained based on the image corresponding to the tree layer, wherein all contours include reference object contours and multiple non-reference object contours. In step S420, the ratio between the pixel width of the reference object contour and the pixel width of each non-reference object contour is obtained. In step S430, the actual diameter at breast height of each non-reference object contour is obtained by comparing the actual diameter at breast height of the reference object with the ratio. In step S440, the average diameter at breast height of the tree layer is obtained by calculating the average value of the actual diameter at breast height of each non-reference object contour.

[0065] In an embodiment of the present application, before obtaining all contours in the image based on the image corresponding to the tree layer, the image corresponding to the tree layer is grayscaled and binarized. The specific grayscale processing and image binarization processing can be referred to the previous text and will not be repeated here.

[0066] In an embodiment of the present application, after all contours including a reference object contour and multiple non-reference object contours are acquired, the pixel width of the reference object contour and the pixel width of each non-reference object contour are acquired, and then the ratio between the pixel width of the reference object contour and the pixel width of each non-reference object contour is calculated.

[0067] In the embodiment of the present application, the actual diameter at breast height of each non-reference object outline is obtained by multiplying the actual diameter at breast height of the reference object by the ratio, that is, the actual diameter at breast height of each tree in the tree layer as a non-reference object is obtained.

[0068] In an embodiment of the present application, the calculation method of the average base diameter of the shrub layer is the same as the calculation method of the average breast diameter of the tree layer. As long as the pixel width of the reference object and the pixel width of the non-reference object contour in the image corresponding to the shrub layer are obtained, the average base diameter of the shrub layer can be obtained based on the actual base diameter of the reference object. This application will not go into details here.

[0069] In the embodiment of the present application, the specific process of calculating the canopy density of the tree layer, the canopy density of the shrub layer, and the canopy density of the herb layer can be found in Figure 5 .

[0070] Figure 5 An exemplary flow chart for calculating canopy density according to an embodiment of the present application is shown.

[0071] like Figure 5 As shown, in step S510, the image corresponding to the level for which the canopy density is to be calculated is subjected to grayscale processing, contrast enhancement processing and image binarization processing in sequence. In step S520, the ratio of the number of white pixels to the total number of pixels in the image subjected to grayscale processing, contrast enhancement processing and image binarization processing is calculated to obtain the canopy density of the corresponding level.

[0072] Specifically, the grayscale processing process and the image binarization processing process can be referred to in the previous text and will not be described in detail here.

[0073] In the embodiment of the present application, the parameter information at different levels also includes the moisture content of the combustibles at different levels. The specific process of calculating the moisture content of the combustibles at different levels in each forest area can be found in Figure 6 .

[0074] Figure 6 An exemplary flow chart for calculating the moisture content according to an embodiment of the present application is shown.

[0075] like Figure 6As shown, in step S610, the images corresponding to different levels of each forest area are converted into HSV color space. In step S620, the area that meets the set HSV threshold range in the HSV color space is extracted. In step S630, the ratio of the extracted area pixels to the total pixels of the corresponding image is calculated, and the moisture content of the combustible material of the corresponding level is obtained based on the ratio.

[0076] In the embodiment of the present application, the HSV threshold range is the HSV threshold range corresponding to green. When the area that meets the set HSV threshold range in the HSV color space is extracted, the green areas in the image corresponding to different levels are extracted. After the green area is extracted, the green degree is measured by calculating the ratio of green area pixels to total pixels. Then, according to the relationship between moisture content and green degree, the moisture content of the combustible material at the corresponding level is obtained.

[0077] In the embodiment of the present application, the relationship between the moisture content and the green degree is: y=-0.6405x+0.5372, where x is the green degree and y is the moisture content.

[0078] After executing step S120, in step S130, biomass information of different levels in each forest area is calculated based on parameter information sets of different levels in each forest area using biomass models of different levels.

[0079] In an embodiment of the present application, biomass information at different levels includes: trunk biomass of tree layer, branch biomass of tree layer, leaf biomass of tree layer, branch biomass of shrub layer, leaf biomass of shrub layer, biomass of herbaceous layer, biomass of surface litter and humus layer.

[0080] After obtaining the biomass information of different levels in each forest area, in step S140, the combustible load of different levels in each forest area is obtained based on the biomass information of different levels in each forest area.

[0081] In the embodiment of the present application, the first calculation formula is used to obtain the combustible load of the tree layer in each area of ​​the forest based on the tree trunk biomass, tree branch biomass, and tree leaf biomass. The first calculation formula is: M1 = [W 乔干 +W 乔枝 +W 乔叶 ]×n1×(1-P1), where W 乔干 is the trunk biomass of the tree layer, W 乔干 =2.52+0.12(D 2 ×h1), W 乔枝 is the tree layer branch biomass, W 乔枝 =0.06(D 2 ×h1) 0.73 , W乔叶 is the leaf biomass of tree layer, W 乔叶 =17.459×h1 0.482 , h1 is the average height of the tree layer, D is the average breast diameter of the tree layer, n1 is the number of trees in the tree layer, and P1 is the moisture content of the tree layer.

[0082] Specifically, W 乔干 =2.52+0.12(D 2 ×h1)、W 乔枝 =0.06(D 2 ×h1) 0.73 and W 乔叶 =17.459×h1 0.482 It is the biomass model of the tree layer.

[0083] Specifically, the number of trees in the aforementioned tree layer can be obtained by performing target recognition on the image corresponding to the segmented tree layer, which will not be elaborated in this application.

[0084] In the embodiment of the present application, the second calculation formula is used to obtain the combustible load of the shrub layer in each area of ​​the forest based on the shrub layer branch biomass and shrub layer leaf biomass. The second calculation formula is: M2 = [W 灌枝 +W 灌叶 ]×S 灌 ×U 灌 ×(1-P2), where W 灌枝 is the shrub layer branch biomass, W 灌枝 =0.28(B 2 ×h2) 0.74 , W 灌叶 is the leaf biomass of the shrub layer, W 灌叶 =10.463×h2 0.476 , h2 is the average height of the shrub layer, B is the average base diameter of the shrub layer, S 灌 is the area of ​​the shrub layer, U 灌 is the density of the shrub layer, and P2 is the moisture content of the shrub layer.

[0085] Specifically, W 灌枝 =0.28(B 2 ×h2) 0.74 and W 灌叶 =10.463×h2 0.476 Biomass model for the shrub layer.

[0086] Specifically, the area of ​​the aforementioned bush layer can be obtained by direct user input or calling map information, etc., and this application does not limit this.

[0087] In the embodiment of the present application, the third calculation formula is used to obtain the fuel load of the herbaceous layer in each area of ​​the forest based on the herbaceous layer biomass. The third calculation formula is: M3 = W 草 ×(1-P3),W 草 is the herbaceous layer biomass, W 草 =0.40×[(S 草 ×U 草 ) 2 ×h3] 0.56 , S 草 is the area of ​​the herbaceous layer, U 草 is the canopy density of the herb layer, h3 is the average height of the herb layer, and P3 is the moisture content of the herb layer.

[0088] Specifically, W 草 =0.40×[(S 草 ×U 草 ) 2 ×h3] 0.56 This is the biomass model of the herbaceous layer.

[0089] Specifically, the area of ​​the aforementioned herbaceous layer can be obtained by direct user input or calling map information, etc., and this application does not limit this.

[0090] In the embodiment of the present application, the fourth calculation formula is used to obtain the combustible load of the surface litter and humus layer in each area of ​​the forest based on the biomass of the surface litter and humus layer. The fourth calculation formula is: M4=W 总 ×S 总 ×P4,W 总 is the biomass of surface litter and humus layer, W 总 =h 腐 +h 地 ×t,h 腐 is the thickness of humus, h 地 is the thickness of the litter on the ground, t is the number of years, S 总 is the area of ​​the surface litter and humus layer, and P4 is the moisture content of the surface litter and humus layer.

[0091] Specifically, W 总 =h 腐 +h 地 ×t is the biomass model of surface litter and humus layer.

[0092] Specifically, the area of ​​the aforementioned surface litter and humus layer can be obtained by direct user input or calling map information, etc., and this application does not limit this.

[0093] Specifically, the thickness of the aforementioned humus and the thickness of the surface litter can be obtained through actual measurements by the user, etc., and this application does not limit this.

[0094] After obtaining the combustible loads at different levels in each forest area, in step S150, the combustible loads in each forest area are calculated based on the combustible loads at different levels in each forest area.

[0095] In an embodiment of the present application, the combustible loads of different levels in various forest areas are added together to obtain the combustible loads of various forest areas.

[0096] In the embodiment of the present application, after the combustible load of each forest area is obtained, the forest fire hazard level can be determined according to the combustible load of each forest area. For example, the area of ​​each forest area is compared with 1ha, and the area of ​​each forest area is converted into 1ha in proportion, so as to obtain the forest combustible load in 1ha of each forest area. Then, according to the range of the forest combustible load in 1ha, the corresponding forest fire hazard level is obtained.

[0097] In one embodiment of the present application, in an area of ​​a forest, a first image is obtained by photographing a treetop at a height of 30 degrees at a height of 1.5 meters from the ground so that the treetop is exactly at the top edge of the photo. A second image is obtained by photographing at a position 5-10 degrees to the left of the first image photographing position, and a third image is obtained by photographing at a position 5-10 degrees to the right of the first image photographing position. Then, the forest combustible load estimation method 100 is used to process the first image, the second image, and the third image in sequence, and the forest combustible load estimation is performed based on the processing results.

[0098] Specifically, the first image captured is as follows: Fig. 7A As shown. By segmenting the first image, we can obtain Figure 7B The image corresponding to the tree layer shown, Figure 7C The image corresponding to the shrub layer shown, Fig.7D The corresponding image of the herb layer shown and Fig. 7E The images shown correspond to the surface litter and humus layers.

[0099] Next, input the reference object height, reference object breast diameter, reference object base diameter and other data in the program command line. The data of each layer corresponding to the first image, the second image and the third image are calculated as follows:

[0100] Tree layer: average tree height: h1 = 3.892 meters; average breast diameter: D = 15.62 centimeters; canopy density: U 乔 =0.32; green degree: 0.32, moisture content: P1 = 0.30;

[0101] Shrub layer: average tree height: h2 = 0.385 m; average base diameter: B = 1.55 cm; canopy density: U灌 =0.29; green degree: 0.43, moisture content: P2 = 0.38;

[0102] Herbaceous layer: average height: h3 = 0.052 cm; canopy density: U 草 =0.71, green degree: 0.26, moisture content: P3 = 0.24;

[0103] Surface litter and humus layer: Biomass of surface litter and humus layer: W 总 =0.02.

[0104] Then, the combustible loads of each layer corresponding to the first image are obtained: M1=776.032, M2=136.604, M3=7.648 and M4=1.49. The combustible loads of each layer corresponding to the second image and the third image are obtained.

[0105] Then, the combustible load of the forest area obtained from the first image is obtained: M 1 =M1+M2+M3+M4=921.78. And the combustible load M of the forest area obtained according to the second image is obtained 2 =921.56, and the combustible load M of the forest area obtained according to the third image is obtained 3 =921.89.

[0106] According to the combustible load M in the forest area, 1 +M 2 +M 3 ) / 3=921.74.

[0107] Finally, the area of ​​the forest area was compared with 1ha, and the area of ​​the shooting area was converted into 1ha in proportion. The area in the figure is about 110m 2 , then the fuel load M is converted to M′=921.74 / 110×10000=8.37 t / ha, which corresponds to a low forest fire risk. Therefore, the forest area is a low-risk area.

[0108] In summary, through the forest combustible load calculation scheme provided above, the embodiment of the present application obtains images corresponding to different levels of each forest area by segmenting the images corresponding to each area of ​​the forest. The weight of combustibles in each layer can be calculated according to the characteristics of different plant levels, and the weight of combustibles in a certain area of ​​the forest can be obtained by cumulative averaging. It will not be affected by environmental factors, nor will it be affected by the canopy layer for the acquisition of data at other levels, and it takes a short time and a small workload.

[0109] The embodiment of the present application also provides a forest combustible load calculation system, which can use the aforementioned forest combustible load calculation method 100 to calculate the forest combustible load, and can also use other forest combustible load calculation methods to calculate the forest combustible load, and the present application does not limit this.

[0110] Figure 8 An exemplary structural block diagram of the forest fuel load measurement system of an embodiment of the present application is shown.

[0111] like Figure 8 As shown, the device 800 includes an image processing module 810, a parameter acquisition module 820, a biomass calculation module 830, and a combustible load calculation module 840. In the embodiment of the present application, the image processing module 810, the parameter acquisition module 820, the biomass calculation module 830, and the combustible load calculation module 840 can be separate units or integrated in an integrated circuit, and the present application does not limit this.

[0112] Specifically, the image processing module 810 is used to segment the images corresponding to each forest area according to the RGB range of forest combustibles at different levels, so as to obtain images corresponding to different levels of each forest area.

[0113] Specifically, the parameter acquisition module 820 is used to calculate parameter information of different levels of each forest area based on images corresponding to different levels of each forest area.

[0114] Specifically, the biomass calculation module 830 is used to calculate the biomass information of different levels in each forest area based on the parameter information sets of different levels in each forest area by using biomass models of different levels.

[0115] Specifically, the combustible load calculation module 840 is used to obtain the combustible load at different levels in each forest area based on the biomass information at different levels in each forest area, and calculate the combustible load in each forest area based on the combustible load at different levels in each forest area.

[0116] When the device 800 uses the aforementioned forest fuel load calculation method 100 to calculate the forest fuel load, the image processing module 810 performs the aforementioned step S110, the parameter acquisition module 820 performs the aforementioned step S120, the biomass calculation module 830 performs the aforementioned step S130, and the fuel load calculation module 840 performs the aforementioned steps S140 and S150. The specific execution process can be referred to above, and will not be repeated here.

[0117] Although multiple embodiments of the present application have been shown and described herein, it is obvious to those skilled in the art that such embodiments are provided only by way of example. Those skilled in the art can think of many changes, modifications and alternatives without departing from the thought and spirit of the present application. It should be understood that in the process of practicing the present application, various alternatives to the embodiments of the present application described herein can be adopted. The attached claims are intended to limit the scope of protection of the present application, and therefore cover equivalents or alternatives within the scope of these claims.

Claims

1. A method for calculating forest fuel load, characterized in that: include: According to the RGB range of forest combustibles at different levels, the images corresponding to each forest area are segmented to obtain images corresponding to different levels of each forest area; Calculate the parameter information of different levels of each forest area based on the images corresponding to different levels of each forest area; Different levels of biomass models are used to calculate the biomass information of different levels in each forest area based on the parameter information sets of different levels in each forest area; Obtain the fuel load at different levels in each forest area based on the biomass information at different levels in each forest area; The fuel load in each forest area is calculated based on the different levels of fuel load in each forest area.

2. The method for calculating forest fuel load according to claim 1, characterized in that: The different layers include the tree layer, the shrub layer, the herb layer, and the surface litter and humus layer.

3. The method for calculating forest fuel load according to claim 2, characterized in that: The parameter information at different levels includes: the average height of the tree layer, the average breast diameter of the tree layer, the number of trees in the tree layer, the canopy density of the tree layer, the average height of the shrub layer, the average base diameter of the shrub layer, the canopy density of the shrub layer, the average height of the herb layer and the canopy density of the herb layer.

4. The method for calculating forest fuel load according to claim 3, characterized in that: In the process of calculating the average height of the tree layer, the average height of the shrub layer, and the average height of the herb layer, the following steps are performed: The images corresponding to the layers whose average height needs to be calculated are processed in grayscale and binarized in sequence; Find all contours in the grayscaled and binarized image; Use preset indexes to match the reference object contours among all contours; Get the pixel height of the reference object outline; Based on the pixel height of the reference object contour, the average height corresponding to the corresponding layer is obtained.

5. The method for calculating forest fuel load according to claim 4, characterized in that: The average height corresponding to the corresponding layer is obtained based on the pixel height of the reference object contour, including: Traverse all contours except the reference object contour to obtain pixel heights of the contours except the reference object contour; Calculate the ratio between the pixel height of each other contour and the pixel height of the reference object contour; The actual height of each other contour is obtained by comparing the actual height of the reference object with the ratio; The average height corresponding to the corresponding level is obtained by calculating the average of the actual heights of each other contour.

6. The method for calculating forest fuel load according to claim 3, characterized in that: In the process of calculating the average DBH of the tree layer, the following steps are performed: Acquire all contours in the image based on the image corresponding to the tree layer, wherein all contours include reference object contours and multiple non-reference object contours; Obtaining the ratio between the pixel width of the reference object contour and the pixel width of each non-reference object contour; The actual breast diameter of each non-reference object contour is obtained by comparing the actual breast diameter of the reference object with the ratio; The average DBH of the tree layer was obtained by calculating the average of the actual DBH of each non-reference object outline.

7. The method for calculating forest fuel load according to claim 3, characterized in that: In the process of calculating the canopy density of the tree layer, the canopy density of the shrub layer, and the canopy density of the herb layer, the following steps are performed: The images corresponding to the level for which the canopy density needs to be calculated are processed in sequence with grayscale, contrast enhancement and image binarization. The ratio of the number of white pixels to the total number of pixels in the image after grayscale processing, contrast enhancement processing and image binarization processing was calculated to obtain the canopy density of the corresponding level.

8. The method for calculating forest fuel load according to claim 3, characterized in that: The parameter information at different levels also includes the moisture content of the fuel at different levels. In the process of calculating the moisture content of the fuel at different levels in each area of ​​the forest, the following steps are performed: Convert the images corresponding to different levels of each forest area to HSV color space; Extract the area in the HSV color space that meets the set HSV threshold range; The ratio of the extracted area pixels to the total pixels of the corresponding image is calculated, and the moisture content of the combustible material at the corresponding layer is obtained based on the ratio.

9. The method for calculating forest fuel load according to claim 3, characterized in that: The biomass information at different levels includes: tree trunk biomass at the tree layer, tree branch biomass at the tree layer, tree leaf biomass at the tree layer, shrub branch biomass at the shrub leaf biomass at the shrub layer, herbaceous layer biomass, and surface litter and humus layer biomass; The first calculation formula is used to obtain the combustible load of the tree layer in each area of ​​the forest based on the tree trunk biomass, tree branch biomass, and tree leaf biomass. The first calculation formula is: M1 = [W 乔干 +W 乔枝 +W 乔叶 ]×n1×(1-P1), where W 乔干 is the trunk biomass of the tree layer, W 乔干 =2.52+0.12(D 2 ×h1), W 乔枝 is the tree layer branch biomass, W 乔枝 =0.06(D 2 ×h1) 0.73 , W 乔叶 is the leaf biomass of tree layer, W 乔叶 =17.459×h1 0.482 , h1 is the average height of the tree layer, D is the average breast diameter of the tree layer, n1 is the number of trees in the tree layer, and P1 is the moisture content of the tree layer; The second calculation formula is used to obtain the fuel load of the shrub layer in each area of ​​the forest based on the shrub layer branch biomass and shrub layer leaf biomass. The second calculation formula is: M2 = [W 灌枝 +W 灌叶 ]×S 灌 ×U 灌 ×(1-P2), where W 灌枝 is the shrub layer branch biomass, W 灌枝 =0.28(B 2 ×h2) 0.74 , W 灌叶 is the leaf biomass of the shrub layer, W 灌叶 =10.463×h2 0.476 , h2 is the average height of the shrub layer, B is the average base diameter of the shrub layer, S 灌 is the area of ​​the shrub layer, U 灌 is the density of the shrub layer, and P2 is the moisture content of the shrub layer; The third calculation formula is used to obtain the fuel load of the herbaceous layer in each area of ​​the forest based on the herbaceous layer biomass. The third calculation formula is: M3 = W 草 ×(1-P3),W 草 is the herbaceous layer biomass, W 草 =0.40×[(S 草 ×U 草 ) 2 ×h3] 0.56 , S 草 is the area of ​​the herbaceous layer, U 草 is the canopy density of the herb layer, h3 is the average height of the herb layer, and P3 is the moisture content of the herb layer; The fourth calculation formula is used to obtain the combustible load of the surface litter and humus layer in each area of ​​the forest based on the biomass of the surface litter and humus layer. The fourth calculation formula is: M4 = W 总 ×S 总 ×P4,W 总 is the biomass of surface litter and humus layer, W 总 =h 腐 +h 地 ×t,h 腐 is the thickness of humus, h 地 is the thickness of the litter on the ground, t is the number of years, S 总 is the area of ​​the surface litter and humus layer, and P4 is the moisture content of the surface litter and humus layer.

10. A forest fuel load calculation system, characterized in that: The forest fuel load calculation method according to any one of claims 1 to 9 is used to calculate the forest fuel load, and the system comprises: An image processing module is used to segment the images corresponding to different areas of the forest according to the RGB range of different levels of forest combustibles, so as to obtain images corresponding to different levels of different areas of the forest; A parameter acquisition module is used to calculate parameter information of different levels of each forest area based on images corresponding to different levels of each forest area; A biomass calculation module is used to calculate the biomass information of different levels in each forest area based on parameter information sets of different levels in each forest area by using biomass models of different levels; The combustible load calculation module is used to obtain the combustible load at different levels in each forest area based on the biomass information at different levels in each forest area, and calculate the forest combustible load of the entire forest area based on the combustible load at different levels in each forest area.

Images