Vegetation coverage observation method based on multiple exposure of phenological camera

Abstract

The invention belongs to the technical field of ecological remote sensing, and discloses a vegetation coverage observation method based on multiple exposure of a phenological camera. Comprising the following steps: S1, shooting a normal exposure image through a phenological camera; s2, setting an exposure compensation value, and respectively shooting two images with different exposure degrees so as to obtain three images with different exposure degrees; s3, normalizing and carrying out weighted fusion on the three images with different exposure degrees so as to obtain an image after exposure fusion; s4, calculating a visible light atmospheric impedance index based on the exposed and fused image, and extracting a vegetation region in the exposed and fused image according to the fact that the visible light atmospheric impedance index is greater than zero; s5, calculating the vegetation coverage through the vegetation area in the image after exposure fusion; according to the method, the vegetation details and textures under the shadow and on the sunny side in the image can be enhanced, so that the background and vegetation in the image can be separated more clearly, and the error of vegetation coverage extraction is reduced.

Description

Technical Field

[0001] The present invention relates to the field of ecological remote sensing technology, and more specifically, to a vegetation coverage observation method based on multiple exposures of a phenological camera. Background Art

[0002] Fractional vegetation cover (FVC) is an important indicator for measuring the distribution and density of surface vegetation. It not only reflects the growth status of vegetation in a region, but is also closely related to ecological functions such as soil protection, water cycle, carbon sequestration, and climate regulation. It is widely used in ecological environment monitoring and assessment. Vegetation with high coverage can effectively prevent soil erosion, reduce greenhouse gas emissions, and maintain biodiversity. It has important application value in ecological research, climate change monitoring, and agricultural management.

[0003] At present, the observation methods of vegetation coverage include manual observation and remote sensing technology observation. Manual observation usually obtains data based on ground surveys and visual estimates, and is suitable for small-scale or high-precision monitoring; remote sensing technology observation is based on satellites, drones and other platforms to obtain optical images, radar images and lidar data, etc., to achieve large-scale, long-term and automatic all-weather monitoring; lidar in near-ground optical remote sensing can provide accurate three-dimensional data, which is suitable for monitoring complex terrain such as forests, but its cost is relatively high; fisheye cameras or phenological cameras can also be used in forests and tall vegetation to observe the vegetation canopy upwards, and calculate the coverage information by separating it from the sky background. This method is low-cost and relatively accurate. The above diversified observation methods have improved monitoring efficiency and at the same time enhanced the early warning capabilities of ecological and environmental changes;

[0004] In the existing near-ground vegetation coverage observation, low-cost phenological cameras are one of the commonly used FVC observation methods. However, for some low vegetation, it is impossible to adopt a vertical upward observation method. It can only be observed vertically downward, and the soil is used as the background. The vegetation coverage is calculated based on threshold segmentation or green channel information. The threshold segmentation algorithm is an algorithm that automatically determines the image threshold. It can automatically select the optimal threshold when the image is binarized to separate the foreground (vegetation) and the background (non-vegetation); the green channel information uses image processing technology to analyze the pixel ratio of the green area in the image to estimate the growth status and coverage of the vegetation. This method usually relies on the contrast between the green part of the plant and the background to achieve accurate separation of vegetation and non-vegetation;

[0005] However, the threshold segmentation algorithm has the following defects: the grayscale histogram-based segmentation method selects the optimal threshold only by the difference in pixel brightness intensity, which makes it easy to be interfered by other ground objects in scenes with complex backgrounds or large brightness changes, resulting in segmentation errors;

[0006] In addition, the green channel information extraction relies on the green characteristics of vegetation, which reduces the influence of non-vegetation to a certain extent. Therefore, in the case of uneven lighting, the illumination part may also cause the color channel of vegetation reflection to deviate due to the spectral composition of ambient light (warm light in the evening) and the reflection characteristics of vegetation, resulting in a possible decrease in the accuracy of the green channel method.

[0007] At the same time, the shadow part will also cause the green information of vegetation to be reduced or missing, and the texture details will become darker or even completely black. In this case, neither green pixel extraction nor threshold segmentation can effectively identify vegetation.

[0008] In view of this, the present invention proposes a vegetation coverage observation method based on multiple exposures of a phenological camera to solve the above problems. Summary of the invention

[0009] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a vegetation coverage observation method based on multiple exposures of a phenological camera, comprising the following steps:

[0010] S1, take a normal exposure image with the phenological camera;

[0011] S2, setting an exposure compensation value, and taking two images with different exposures respectively, thereby obtaining three images with different exposures;

[0012] S3, normalizing and weighted fusion of three images with different exposures to obtain an exposure fused image;

[0013] S4, calculating the visible light atmospheric impedance index based on the image after exposure fusion, and extracting the vegetation area in the image after exposure fusion according to the visible light atmospheric impedance index being greater than zero;

[0014] S5. Calculate vegetation coverage by exposing the vegetation area in the fused image.

[0015] Furthermore, the step of taking a normally exposed image by a phenological camera includes:

[0016] Choose a phenological camera that supports automatic exposure;

[0017] The built-in metering system of the phenology camera automatically adjusts the exposure value and takes a normally exposed image.

[0018] Furthermore, the step of setting the exposure compensation value and taking two images with different exposures respectively includes:

[0019] Set the exposure compensation value of the phenology camera to +2EV and take a long exposure image;

[0020] Set the exposure compensation value of the phenology camera to -1EV and take a short exposure image;

[0021] Then, two images with different exposures are obtained.

[0022] Furthermore, the step of normalizing and weighted fusion of three images with different exposures to obtain an exposure fused image includes:

[0023] Normalizing the pixel brightness of the normal exposure image, the long exposure image, and the short exposure image;

[0024] Subsequently, the three normalized images with different exposure levels are weighted fused to obtain an exposure fused image.

[0025] Furthermore, the method of performing weighted fusion includes:

[0026] Calculate the weight of each pixel in three images with different exposures respectively, and the expression is:

[0027]

[0028] Where Wi is the weight of each pixel in the i-th image, and Ii is the normalized pixel brightness of the corresponding pixel in the i-th image;

[0029] For each pixel of the three images with different exposures, weighted fusion is performed according to the weight, and the expression is:

[0030]

[0031] Where O(x, y) is the pixel of the exposure fused image at position (x, y), Wi(x, y) is the weight of the i-th image at position (x, y), and Ii(x, y) is the normalized pixel brightness of the i-th image at position (x, y).

[0032] Furthermore, the step of calculating the visible light atmospheric impedance index based on the image after exposure and fusion, and extracting the vegetation area in the image after exposure and fusion according to the visible light atmospheric impedance index being greater than zero, includes:

[0033] The expression for calculating the visible light atmospheric impedance index based on the exposure fusion image is set as:

[0034]

[0035] Where VAR I represents the visible atmospheric impedance index, which can indicate the green area of ​​vegetation and highlight the difference of vegetation in the visible light part, G represents the green channel, R represents the red channel, and B represents the blue channel;

[0036] The vegetation area can be obtained by extracting the area where the visible light atmospheric impedance index is greater than zero in the exposure fused image.

[0037] Furthermore, the step of calculating the vegetation coverage by exposing the vegetation area in the fused image includes:

[0038] The number of pixels corresponding to the vegetation area in the exposure fused image is set to P VARI>0 , the total number of pixels of the image after exposure fusion is set to P total ;

[0039] Then, the expression of vegetation coverage corresponding to the image after exposure fusion is:

[0040]

[0041] Where FVC represents vegetation coverage.

[0042] The technical effects and advantages of the vegetation coverage observation method based on multiple exposures of a phenological camera of the present invention are as follows:

[0043] By taking three images with different exposures through three exposure sampling, the vegetation reflection information of the shadow and the sunny side (bright side) in the image can be better identified, which is of great significance for the coverage extraction in the vegetation observation scene with soil as the background; compared with the single automatic exposure method, the present invention can accurately identify the shadow and the sunny side (bright side), reduce the lack of channel signals in the shadow and the underestimation caused by the saturation of the sunny side signal, and improve the accuracy of vegetation coverage observation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a vegetation coverage observation method based on multiple exposures of a phenological camera according to the present invention;

[0045] Figure 2 is a schematic diagram of three images with different exposures and an image after exposure fusion in the present invention;

[0046] Figure 3 A schematic diagram of VARI calculation for an image after exposure fusion in the present invention;

[0047] Figure 4 The vegetation area indicated by the VARI image of the image after exposure and fusion in the present invention;

[0048] Figure 5 Schematic diagram of comparison of vegetation coverage after green channel threshold segmentation of the normal exposure image (left) and vegetation coverage after VARI image indication of vegetation area of ​​the exposure fused image (right) in the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Example 1

[0051] See also Figures 1 to 5 As shown, the vegetation coverage observation method based on multiple exposures of a phenological camera described in this embodiment includes the following steps:

[0052] S1, take a normal exposure image with the phenological camera;

[0053] S2, setting an exposure compensation value, and taking two images with different exposures respectively, thereby obtaining three images with different exposures (a normal exposure image, a long exposure image, and a short exposure image);

[0054] S3, normalizing and weighted fusion of three images with different exposures to obtain an exposure fused image;

[0055] S4, calculating the visible light atmospheric impedance index based on the image after exposure fusion, and extracting the vegetation area (vegetation area information) in the image after exposure fusion according to the visible light atmospheric impedance index being greater than zero;

[0056] S5. Calculate vegetation coverage by exposing the vegetation area in the fused image.

[0057] Furthermore, the step of taking a normally exposed image by a phenological camera includes:

[0058] S11. Select a phenological camera that supports the automatic exposure (AE) function;

[0059] S12. The built-in metering system of the phenology camera automatically adjusts the exposure value (EV) and takes a normally exposed image.

[0060] Specifically, the phenology camera supports the automatic exposure (AE) function. When shooting, the camera's built-in metering system analyzes the overall brightness of the scene and automatically adjusts the exposure value (EV) to take a normally exposed image.

[0061] Furthermore, the step of setting the exposure compensation value and taking two images with different exposures respectively includes:

[0062] S21, setting the exposure compensation value of the phenological camera to +2EV, and taking a long exposure image;

[0063] S22, setting the exposure compensation value of the phenological camera to -1EV, and taking a short exposure image;

[0064] S23. Then, two images with different exposures are obtained.

[0065] Specifically, by setting the exposure compensation value, the phenological camera can increase and decrease the exposure based on the automatic exposure calculation. The exposure compensation values ​​are set to +2EV and -1EV respectively, so that long-exposure images and short-exposure images can be captured.

[0066] Furthermore, the step of normalizing and weighted fusion of the three images with different exposures to obtain an exposure fused image includes:

[0067] S31, normalizing the pixel brightness of the normal exposure image, the long exposure image, and the short exposure image;

[0068] S32. Subsequently, weighted fusion is performed on the three normalized images with different exposure levels to obtain an exposure-fused image.

[0069] Specifically, data processing is performed on the three images with different exposures, the pixel brightness of the three images with different exposures is normalized, and weighted fusion is performed to maximize the retention of medium brightness information and remove pixels that are too dark or too bright.

[0070] Furthermore, the weighted fusion method includes:

[0071] Calculate the weight of each pixel in three images with different exposures respectively, and the expression is:

[0072]

[0073] Where Wi is the weight of each pixel in the i-th image, and Ii is the normalized pixel brightness of the corresponding pixel in the i-th image;

[0074] For each pixel of the three images with different exposures, weighted fusion is performed according to the weight, and the expression is:

[0075]

[0076] Where O(x, y) is the pixel value of the exposure fused image at position (x, y), Wi(x, y) is the weight of the i-th image at position (x, y), and Ii(x, y) is the normalized pixel brightness of the i-th image at position (x, y).

[0077] Further, the step of calculating the visible light atmospheric impedance index based on the image after exposure fusion, and extracting the vegetation area in the image after exposure fusion according to the visible light atmospheric impedance index being greater than zero includes:

[0078] S41, setting the expression for calculating the visible light atmospheric impedance index based on the image after exposure fusion to:

[0079]

[0080] In the formula, VARI represents the visible atmospheric impedance index, which can indicate the green area of ​​vegetation and highlight the difference of vegetation in the visible light part, G represents the green channel, R represents the red channel, and B represents the blue channel;

[0081] S42. Extract the area where the visible light atmospheric impedance index is greater than zero in the exposure fused image to obtain the vegetation area.

[0082] Furthermore, the step of calculating the vegetation coverage by exposing the vegetation area in the fused image includes:

[0083] S51, setting the number of pixels corresponding to the vegetation area in the image after exposure fusion to P VARI>0 , the total number of pixels of the image after exposure fusion is set to P total , then, the expression of vegetation coverage corresponding to the image after exposure fusion is:

[0084]

[0085] Where FVC represents vegetation coverage.

[0086] It should be noted that by taking vegetation images in sequence with different exposures, long exposure can obtain the shadow image, and short exposure can obtain the details of the sunny side (bright area). Through normalization and brightness weighted fusion, the medium brightness information in the three pictures (three images with different exposures) can be retained to the maximum extent, and the pixels that are too dark or too bright can be removed. The best brightness of different pixels is selected to synthesize a multi-exposure image, and then the visible atmospheric impedance index (VARI) is calculated. The VARI index is used to highlight the characteristics of vegetation in the visible light part. It has a certain anti-interference ability to atmospheric and light changes. According to the extraction of the area with VARI>0 as the vegetation area, and the coverage is calculated based on this, the vegetation coverage can be accurately calculated;

[0087] Among them, Figure 2 In the figure, ①-③ are three images with different exposures, and ④ is the image after exposure fusion; VARI image ( Figure 3 ), by identifying the area with VARI>0 as the vegetation area, calculate FVC( Figure 4) is 67.36%, compared with the vegetation coverage of 62.36% calculated from a single image taken with automatic exposure ( Figure 5 ), and further, the results show that the calculated vegetation coverage and the separation of vegetation and background in the exposure fused image are more accurate (especially the separation in the vegetation shadow).

[0088] In this embodiment, by acquiring three images with different exposures based on the phenological camera, the reflection information of ground objects under different light conditions and backgrounds can be accurately extracted, thereby realizing accurate calculation of vegetation coverage; the visible atmospheric impedance index (VARI) can be used to segment vegetation and non-vegetation areas, thereby highlighting the characteristics of vegetation in the visible light part, and having a certain anti-interference ability to atmospheric and light changes.

[0089] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0090] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0091] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0092] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A vegetation coverage observation method based on multiple exposures of a phenological camera, characterized in that: The following steps are involved: S1, take a normal exposure image with the phenological camera; S2, setting an exposure compensation value, and taking two images with different exposures respectively, thereby obtaining three images with different exposures; S3, normalizing and weighted fusion of three images with different exposures to obtain an exposure fused image; S4, calculating the visible light atmospheric impedance index based on the image after exposure fusion, and extracting the vegetation area in the image after exposure fusion according to the visible light atmospheric impedance index being greater than zero; S5. Calculate vegetation coverage by exposing the vegetation area in the fused image.

2. The method for observing vegetation coverage based on multiple exposures of a phenological camera according to claim 1, characterized in that: The step of taking a normally exposed image by using a phenological camera comprises: Choose a phenological camera that supports automatic exposure; The built-in metering system of the phenology camera automatically adjusts the exposure value and takes a normally exposed image.

3. The vegetation coverage observation method based on multiple exposures of a phenological camera according to claim 1 is characterized in that: The step of setting the exposure compensation value and taking two images with different exposures respectively includes: Set the exposure compensation value of the phenology camera to +2EV and take a long exposure image; Set the exposure compensation value of the phenology camera to -1EV and take a short exposure image; Then, two images with different exposures are obtained.

4. The method for observing vegetation coverage based on multiple exposures of a phenological camera according to claim 1, characterized in that: The step of normalizing and weighted fusion of three images with different exposures to obtain an exposure fused image includes: Normalizing the pixel brightness of the normal exposure image, the long exposure image, and the short exposure image; Subsequently, the three normalized images with different exposure levels are weighted fused to obtain an exposure fused image.

5. The method for observing vegetation coverage based on multiple exposures of a phenological camera according to claim 4, characterized in that: The method of performing weighted fusion includes: Calculate the weight of each pixel in three images with different exposures respectively, and the expression is: Where Wi is the weight of each pixel in the i-th image, and Ii is the normalized pixel brightness of the corresponding pixel in the i-th image; For each pixel of the three images with different exposures, weighted fusion is performed according to the weight, and the expression is: Where O(x,y) is the pixel of the exposure fused image at position (x,y), Wi(x,y) is the weight of the i-th image at position (x,y), and Ii(x,y) is the normalized pixel brightness of the i-th image at position (x,y).

6. The method for observing vegetation coverage based on multiple exposures of a phenological camera according to claim 1, characterized in that: The step of calculating the visible light atmospheric impedance index based on the image after exposure and fusion, and extracting the vegetation area in the image after exposure and fusion according to the visible light atmospheric impedance index being greater than zero, comprises: The expression for calculating the visible light atmospheric impedance index based on the exposure fusion image is set as: In the formula, VARI represents the visible light atmospheric impedance index, G represents the green channel, R represents the red channel, and B represents the blue channel; The vegetation area can be obtained by extracting the area where the visible light atmospheric impedance index is greater than zero in the exposure fused image.

7. The method for observing vegetation coverage based on multiple exposures of a phenological camera according to claim 6, characterized in that: The step of calculating the vegetation coverage by exposing the vegetation area in the fused image includes: The number of pixels corresponding to the vegetation area in the exposure fused image is set to P VARI>0 , the total number of pixels of the image after exposure fusion is set to P total ; Then, the expression of vegetation coverage corresponding to the image after exposure fusion is: Where FVC represents vegetation coverage.

Images