A vegetation growth monitoring system and method integrating visible light and near-infrared dual cameras

Abstract

The present application relates to a kind of vegetation vigor monitoring method of integrated visible light and near-infrared dual camera, the method comprises: S1, using visible light monitoring camera to collect the RGB image of monitoring area, using near-infrared camera to collect the near-infrared waveband image of monitoring area;S2, utilize boundary search algorithm to obtain the common boundary of RGB image and near-infrared waveband image, and extract the overlapping area image of RGB image and near-infrared waveband image;S3, utilize ground eighteen degrees standard gray card to the color of the overlapping area image is corrected, according to the correction result, the normalized difference vegetation index NDVI image of the overlapping area image is calculated, then combined with the corrected RGB image, the monitoring result of vegetation vigor is obtained.The present application solves the current conventional visible light (RGB) camera monitoring can only artificial experience from ordinary photo qualitative identification vegetation vigor, multispectral monitoring equipment is expensive, highly dependent on professional software, difficult to popularize, and satellite remote sensing monitoring cannot reach field scale and time resolution difference and other problems.

Description

Technical Field

[0001] This invention relates to the field of plant growth information monitoring technology, and in particular to a vegetation growth monitoring system and method integrating visible light and near-infrared dual cameras. Background Technology

[0002] Monitoring and studying the spatiotemporal dynamics of vegetation has significant scientific and applied value, especially monitoring crop growth and trend changes, which can provide important data for field management and yield assessment. Currently, commonly used vegetation monitoring methods include three types: monitoring using red-green-blue (RGB) cameras based on IoT technology, ground-based monitoring using multispectral analyzers, and regional monitoring using satellite remote sensing multispectral images. Conventional RGB camera monitoring only obtains images of crops in the visible light spectrum, requiring manual interpretation of the images to identify crop growth, making it overly empirical and qualitative, and the monitoring results are significantly affected by human factors. Ground-based monitoring using multispectral analyzers can obtain Normalized Difference Vegetation Index (NDVI) images through band calculations. Healthy vegetation (chlorophyll) reflects more near-infrared (NIR) light but absorbs more red light. The NDVI quantifies vegetation growth information by measuring the difference between near-infrared (strongly reflected) and red light (absorbed) light. This parameter value ranges between -1 and +1. A higher NDVI value will result from lower reflectance in the red channel and higher reflectance in the NIR channel, and vice versa. NDVI is a standardized method for measuring vegetation health; a higher NDVI value indicates healthier vegetation. A lower NDVI value indicates poor or no vegetation. In agriculture, vegetation NDVI is often used for precision agriculture and biomass measurement. In forestry, NDVI is used to quantify forest supply and leaf area index. Furthermore, NDVI is a good indicator of drought; when water restricts vegetation growth, the relative NDVI value is lower. In the real world, hundreds of applications are using NDVI for vegetation monitoring.

[0003] NDVI is currently the best spectral index for monitoring vegetation growth and coverage. However, multispectral instruments are expensive, have limited availability and widespread adoption, and require manual data collection, which is time-consuming, labor-intensive, and cannot provide real-time monitoring. Furthermore, this method requires specialized software for spectral data processing, further limiting its applicability. Satellite remote sensing multispectral images are often limited by temporal and spatial resolution; for example, weather conditions significantly affect data usability. Remote sensing data processing also requires specialized software and knowledge, making it technically challenging and limiting its widespread adoption and promotion. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] In view of the shortcomings of the existing technology, the present invention provides a vegetation growth monitoring system and method integrating visible light and near-infrared dual cameras, which solves the technical problems of high cost of existing vegetation growth monitoring equipment, large influence of monitoring results on human experience factors or environmental factors (such as weather), high dependence on professional software / knowledge for data processing, and poor popularization and promotion.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0008] In a first aspect, the present invention provides a method for monitoring vegetation growth by integrating visible light and near-infrared dual cameras, the method comprising:

[0009] S1. A visible light monitoring camera is used to collect RGB images of the monitoring area, and a near-infrared camera is used to collect near-infrared band images of the monitoring area.

[0010] S2. Use the boundary search algorithm to obtain the common boundary between the RGB image and the near-infrared band image, and extract the overlapping area image of the RGB image and the near-infrared band image.

[0011] S3. Use the 18-degree ground standard gray card to correct the color of the overlapping area image, calculate the normalized vegetation index (NDVI) image of the overlapping area image based on the correction result, and then combine it with the corrected RGB image to realize the monitoring of vegetation growth in the monitoring area.

[0012] According to a preferred embodiment of the present invention, in S2, the boundary search algorithm sets the number of deviation pixels to 40 pixels. The search algorithm is based on the principle that the correlation between different bands of the same ground object image is the highest. It calculates the absolute value of the correlation coefficient between the red light band and the near-infrared band for each single pixel of the corresponding 40 rows or columns of the upper, lower, left, and right boundaries. According to the position of the maximum absolute value of the correlation coefficient, the four-sided boundaries of the RGB image and the near-infrared band image are found. The overlapping area of ​​the images is determined according to the four-sided boundaries of the RGB image and the near-infrared band image.

[0013] According to a preferred embodiment of the present invention, in S3, the method for correcting the color of the overlapping region image specifically includes: correcting the red band and calculating the reflectance of the red band image, correcting the near-infrared band and calculating the reflectance of the near-infrared band image, and correcting the red, green and blue band colors in the RGB image.

[0014] The method for correcting the red band and calculating the reflectance of the red band image is as follows:

[0015] Read the average value of all pixels from the red band image of the RGB image using an 18-degree standard grayscale card, and set it to DN. g18 The reflectance r of a red-band image is calculated using the following formula. R :

[0016]

[0017] In the formula, DN R These are the numerical values ​​of each pixel in the red light band;

[0018] The method for correcting the near-infrared band and calculating the reflectance of the near-infrared band image is as follows: The average value of all pixels on an 18-degree standard gray card is read from the near-infrared image and set as DN. g18 The reflectance r of a near-infrared image is calculated using the following formula. NIR :

[0019]

[0020] In the formula, DN NIR These are the values ​​of each pixel in the near-infrared band.

[0021] The method for correcting the red, green, and blue band colors in an RGB image is as follows:

[0022] Red light band:

[0023]

[0024] Green band:

[0025]

[0026] Blue light band:

[0027]

[0028] In the above formula, DN g18R DN g18G DN g18B These represent the average pixel values ​​of all pixels on an 18-degree standard gray card in the R, G, and B bands, respectively, and DN. R0 DN G0 DN B0 N represents the grayscale values ​​of the red, green, and blue bands of the RGB image before correction. R DN G DN B These are the grayscale values ​​of the red, green, and blue bands after RGB image correction.

[0029] According to a preferred embodiment of the present invention, in S3, the method for calculating the Normalized Difference Vegetation Index (NDVI) image is as follows: using the reflectance r of the corrected red band image... RThe reflectance r of near-infrared band images NIR The normalized vegetation index (NDVI) image of the overlapping region was calculated:

[0030]

[0031] Secondly, the present invention provides a vegetation growth monitoring system integrating visible light and near-infrared dual cameras, the system comprising:

[0032] A combination camera system placed at a high position, the combination camera system including a visible light monitoring camera and a near-infrared camera;

[0033] An 18-degree standard gray card is installed on the ground in the monitored area. The 18-degree standard gray card can be captured by the combined camera. The 18-degree standard gray card provides a reference for correction of each band of RGB image and near-infrared image.

[0034] The data acquisition and transmission device acquires RGB images and near-infrared band images of the monitoring area captured by the combined camera, and transmits the acquired data to a remote monitoring computer.

[0035] The system according to the present invention further includes a monitoring pole for mounting the combined camera at a high position; and the monitoring pole is also provided with a solar cell module, which absorbs solar energy and converts it into electrical energy to power the combined camera and the data acquisition and transmission device.

[0036] According to the system of the present invention, the data acquisition and transmission device includes a rechargeable battery, a power management module, a data acquisition and processing module, and a communication antenna; the rechargeable battery is connected to the solar cell module for storing electrical energy; the data acquisition and processing module transmits the acquired data to a remote monitoring computer through the communication antenna.

[0037] According to the system of the present invention, in the combined camera, the effective pixels of the visible light monitoring camera and the near-infrared camera are 1920*1080.

[0038] According to the system of the present invention, the surface of the 18-degree standard gray card is parallel to the camera mirrors of the visible light monitoring camera and the near-infrared camera.

[0039] (III) Beneficial Effects

[0040] This invention integrates a near-infrared (NIR) camera into a conventional IoT-based red, green, and blue (RGB) camera, forming a dual-camera vegetation monitoring device. This device simultaneously photographs vegetation, adding a near-infrared image to the existing visible light (red, green, and blue) images. The NDVI is calculated using the reflectance of the red (R) band in both the NIR and visible light images: NDVI = (NIR - R) / (NIR + R). Because the integrated dual-camera images exhibit spatial inconsistencies, this invention proposes an algorithm based on the highest correlation between the common boundaries of the visible and near-infrared images. This algorithm automatically searches for four common boundaries between the two images, thus identifying the overlapping common area. To eliminate the influence of background light variations at the monitoring site, this invention introduces an 18-degree grayscale standard, which is installed on the monitored ground. The 18-degree grayscale standard is used to correct the reflectance of each band in the captured visible and near-infrared images, ensuring consistency and comparability of NDVI values ​​calculated at different times.

[0041] The present invention solves the technical problems of conventional visible light (RGB) camera monitoring, which can only rely on manual experience to qualitatively identify vegetation growth from ordinary photos; multispectral monitoring equipment is expensive and mostly used for professional monitoring or drone-borne aerial monitoring, and requires professional software for post-processing of data to calculate the NDVI index, resulting in poor popularization and promotion; and satellite remote sensing monitoring has insufficient spatial resolution to reach the field scale and insufficient temporal resolution. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the vegetation growth monitoring system of the present invention.

[0043] Figure 2 This is a schematic diagram of the combined camera of the vegetation growth monitoring system of the present invention.

[0044] Figure 3 Images captured by the visible light monitoring camera and the near-infrared camera in a combined camera system may exhibit the following four types of positional deviations.

[0045] Figure 4 This document outlines the process for determining the left common boundary of images captured by a light monitoring camera and a near-infrared camera using a boundary search algorithm.

[0046] Figure 5 This document outlines the process for determining the right common boundary of images captured by a light monitoring camera and a near-infrared camera using a boundary search algorithm.

[0047] Figure 6 This document outlines a process for calculating the upper common boundary of images captured by a light monitoring camera and a near-infrared camera using a boundary search algorithm.

[0048] Figure 7 This document outlines a process for calculating the lower common boundary of images captured by a light monitoring camera and a near-infrared camera using a boundary search algorithm. Detailed Implementation

[0049] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0050] like Figure 1 The diagram illustrates a preferred embodiment of the vegetation growth monitoring system integrating visible light and near-infrared dual cameras according to the present invention. As shown, the system includes a monitoring pole 3, the lower end of which is embedded in the soil of the monitored vegetation area. A combined camera 20, comprising a visible light monitoring camera 21 and a near-infrared camera 22, is installed at the upper end of the monitoring pole 3. An 18-degree standard gray card 1 is mounted on the ground of the monitoring area via a bracket. The surface of the standard gray card 1 is parallel to the mirror surfaces of the visible light monitoring camera 21 and the near-infrared camera 22, so that the monitoring camera 21 and the near-infrared camera 22 can orthogonally capture images of the gray card. The system also includes a data acquisition and transmission device 40 for acquiring RGB images and near-infrared band images of the monitoring area captured by the combined camera 20, and transmitting the acquired data to a remote monitoring computer.

[0051] Among them, the visible light monitoring camera 21 has a resolution of 2 megapixels and an effective pixel count of 1920*1080; the near-infrared camera 22 has a spectral response range of 0.9um-1.7um, and its image resolution is also 2 megapixels with an effective pixel count of 1920*1080. The 18-degree standard gray card 1 is made of PVC material and is used to calibrate the RGB and near-infrared images acquired by the combined camera 20.

[0052] The monitoring rod 3 is also equipped with a solar panel 5, which absorbs solar energy and converts it into electrical energy to power the combined camera 20 and the data acquisition and transmission device 40. The data acquisition and transmission device 40 includes a rechargeable battery 41, a power management module 42, a data acquisition and processing module 43, and a communication antenna 44. The rechargeable battery 41 is connected to the solar panel 5 and stores electrical energy; it also provides operating power to the data acquisition and processing module 43, the communication antenna 44, and the combined camera 20. The data acquisition and processing module 43 transmits the acquired data to a remote monitoring computer via the communication antenna 44. The data acquisition and processing module 43 is communicatively connected to the combined camera 20. Except for the communication antenna 44, which extends outside the housing, all other modules in the data acquisition and transmission device 40 are enclosed within a housing.

[0053] The method for using the above system to monitor vegetation growth is as follows:

[0054] (1) The RGB image and near-infrared band image of the monitored area are acquired by the visible light monitoring camera 21 and the near-infrared camera 22, respectively;

[0055] (2) The data acquisition and processing module 43 executes the following programs, including:

[0056] Step 1: Use a boundary search algorithm to obtain the common boundary between the RGB image and the near-infrared band image, and extract the overlapping area image between the RGB image and the near-infrared band image.

[0057] Because the RGB camera and the near-infrared camera are not positioned together, the images they capture have positional differences. It is necessary to extract the overlapping area of ​​these two images to calculate the required NDVI index. This invention is based on the principle that different bands of images of the same ground feature have the highest correlation. It uses a search algorithm to obtain the overlapping boundary of these two images, and then obtains the overlapping area image. Images captured by the RGB camera and the near-infrared camera may have the following four types of positional deviations, such as... Figure 3 As shown.

[0058] Because the dual cameras are positioned very close to each other, the captured images have very small positional deviations, typically only a few to a dozen pixels. The boundary search algorithm proposed in this invention sets the number of deviation pixels to 40 pixels to accommodate situations with larger image deviations.

[0059] The search algorithm includes determining the left, upper, lower, and right boundaries. During boundary calculation, based on the principle that different bands of the same ground feature image have the highest correlation, the absolute values ​​of the correlation coefficients between the red and near-infrared bands are calculated for each individual pixel in each row or column corresponding to the upper, lower, left, and right boundaries. Based on the location of the maximum absolute value of the correlation coefficient, the four-sided boundaries of the RGB and near-infrared images are identified, and the overlapping areas of the images are determined based on these boundaries. Specifically, the process for determining the left common boundary is as follows: Figure 4 As shown, the process for determining the right common boundary is as follows: Figure 5 As shown, the process for determining the upper common boundary is as follows: Figure 6 As shown, the process for determining the lower common boundary is as follows: Figure 7 As shown.

[0060] by Figure 4Taking the search process for the left common boundary as an example: Initialize the correlation coefficient r = 0. Read the first (i = 1) column of data in the red band of the RGB image into the R

[1080] array. Read the first (j = 1) column of data in the near-infrared band into NIR

[1080] , calculate the correlation coefficient between R[41:1040] and NIR[41:1040], if it is greater than r, assign it to r, and record the left boundary L1 = i of the RGB image and the left boundary L2 = j of the near-infrared band. Then read the second (j = 2) column of data in the near-infrared band into NIR

[1080] , and compare again until j = 40. Then read the second (i = 2) column of data in the red band of the RGB image into the R

[1080] array and repeat the above comparison search until i = 40. The L1 and L2 values ​​corresponding to the highest correlation coefficient r among these 40*40 combinations are obtained through iteration. Specifically, the left common boundary of the RGB image is column L1, and the left common boundary of the near-infrared band is column L2. The search methods for the right, upper, and lower common boundaries are similar to the above process.

[0061] Specifically: For RGB camera images with overlapping areas, the cropping range is as follows: the top left pixel row and column numbers are U1 and L1, and the bottom right pixel row and column numbers are B1 and R1. For near-infrared camera images, the cropping range is as follows: the top left pixel row and column numbers are U2 and L2, and the bottom right pixel row and column numbers are B2 and R2.

[0062] It should be noted that the process for determining the above-mentioned common boundaries is unrelated to the effective pixels of the visible light monitoring camera 21 and the near-infrared camera 22 used. Cameras with different effective pixels can still use this algorithm to extract pixels in the overlapping area.

[0063] Step 2: Correct the colors of the overlapping area image using the 18-degree grayscale standard.

[0064] The correction includes red light band, near-infrared band, and correction of each band of the RGB image. After obtaining the overlapping area image of the RGB camera and the near-infrared camera, the red light and near-infrared bands are corrected and their reflectivity is calculated.

[0065] ① RGB image red, green, and blue light band correction:

[0066] Due to variations in ambient light, images captured by ordinary RGB cameras at different times and locations often exhibit differences in hue and brightness, affecting the interpretation and comparison of vegetation growth. This invention uses an 18-degree grayscale standard as a reference to perform color correction on the captured images, ensuring comparability between vegetation images captured by the same camera at different times or by different cameras at different locations. The correction method is as follows:

[0067] Red light band:

[0068]

[0069] Green band:

[0070]

[0071] Blue light band:

[0072]

[0073] In the formula, DN g18R DN g18G DN g18B These represent the average pixel values ​​of all pixels on an 18-degree standard gray card in the R, G, and B bands, respectively, and DN. R0 DN G0 DN B0 N represents the grayscale values ​​of the red, green, and blue bands of the RGB image before correction. R DN G DN B These are the grayscale values ​​of the red, green, and blue bands after RGB image correction. Based on the 18-degree standard grayscale chart, the grayscale values ​​of the 18-degree standard grayscale chart in each image are close to 127. The colors of other ground features have been corrected to eliminate the influence of background light, ensuring comparability.

[0074] ② Correction of the red light band and calculation of reflectivity:

[0075] The reflectance of the 18-degree standard gray card in the red band is 0.18. The average value of all pixels of the 18-degree standard gray card is read from the red band image of the RGB image and set as DN. g18 The reflectance (r) of the corrected red band image is calculated using the following formula. R ):

[0076]

[0077] In the formula, DN R These are the values ​​of each pixel in the red light band.

[0078] ③ Correction of near-infrared band and calculation of reflectivity:

[0079] The reflectance of the 18-degree standard gray card in the near-infrared band is 0.18. The average value of all pixels of the 18-degree standard gray card is read from the near-infrared image and set as DN. g18 The reflectance (r) of the corrected near-infrared band image is calculated using the following formula. NIR ):

[0080]

[0081] In the formula, DN NIR These are the values ​​of each pixel in the near-infrared band.

[0082] Step 3: Calculate the Normalized Difference Vegetation Index (NDVI) image of the overlapping region based on the correction results.

[0083] Using the reflectance (r) of the corrected red band image R The reflectance (r) of near-infrared band images NIR The Normalized Difference Vegetation Index (NDVI) image of the monitored image was calculated:

[0084]

[0085] In monitoring plant growth in the monitored area, this invention integrates and expands upon conventional image monitoring by incorporating a near-infrared camera. Through an algorithm, visible light and near-infrared images are spatially registered, and corrected using an 18-degree grayscale standard to calculate the Normalized Difference Vegetation Index (NDVI), commonly used for vegetation monitoring. This adds an NDVI image to the conventional image information, while simultaneously color correcting the original RGB images to eliminate background light interference, ensuring comparability and effectively improving the monitoring capability of vegetation growth. The entire system is rationally designed and employs a scientific algorithm, solving the problem of conventional cameras being unable to monitor vegetation growth and achieving long-term, accurate, online monitoring of vegetation growth.

[0086] Compared with the prior art, the present invention has the following advantages:

[0087] (1) By combining a near-infrared camera with a conventional RGB camera, an algorithm for obtaining images of the overlapping area of ​​the two cameras was proposed. This provides red light and near-infrared band information for the calculation of the NDVI index, enabling low-cost acquisition of the vegetation NDVI index and online monitoring of vegetation growth.

[0088] (2) The 18-degree standard gray card was used to correct the reflectivity of red light and near-infrared bands, thereby obtaining and improving the accuracy of the vegetation NDVI index.

[0089] (3) Using the 18-degree standard gray card as a benchmark, the color correction of the red, green and blue bands in conventional RGB camera images is improved, thus enhancing the comparability of vegetation monitoring images.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring vegetation growth by integrating visible light and near-infrared dual cameras, characterized in that, The method includes: S1. A visible light monitoring camera is used to collect RGB images of the monitoring area, and a near-infrared camera is used to collect near-infrared band images of the monitoring area. S2. Use the boundary search algorithm to obtain the common boundary between the RGB image and the near-infrared band image, and extract the overlapping area image of the RGB image and the near-infrared band image. The boundary search algorithm sets the number of deviation pixels to 40 pixels. The search algorithm is based on the principle that the correlation between different bands of the same ground object image is the highest. It calculates the absolute value of the correlation coefficient between the red light band and the near-infrared band for each individual pixel of the corresponding 40 rows or columns of the upper, lower, left, and right boundaries. According to the position of the maximum absolute value of the correlation coefficient, the four-sided boundaries of the RGB image and the near-infrared band image are found. The overlapping area of ​​the images is determined according to the four-sided boundaries of the RGB image and the near-infrared band image. S3. Use the 18-degree ground standard gray card to correct the color of the overlapping area image, calculate the normalized vegetation index (NDVI) image of the overlapping area image based on the correction result, and then combine it with the corrected RGB image to realize the monitoring of vegetation growth in the monitoring area. The method for color correction of the overlapping region image is as follows: correcting the red band and calculating the reflectance of the red band image, correcting the near-infrared band and calculating the reflectance of the near-infrared band image, and correcting the red, green, and blue band colors in the RGB image; wherein, the method for correcting the red band and calculating the reflectance of the red band image is as follows: Read the average value of all pixels from the red band image of the RGB image using an 18-degree standard grayscale card, and set it to... The reflectance r of a red-band image is calculated using the following formula. R : In the formula, These are the numerical values ​​of each pixel in the red light band; The method for correcting the near-infrared band and calculating the reflectance of near-infrared images is as follows: The average value of all pixels on an 18-degree standard gray card is read from the near-infrared image and set as... The reflectance r of a near-infrared image is calculated using the following formula. NIR : In the formula, These are the values ​​of each pixel in the near-infrared band; The method for correcting the red, green, and blue band colors in an RGB image is as follows: Red light band: Green band: Blue light band: In the above formula, , , These represent the average values ​​of all pixels on an 18-degree standard gray card in the R, G, and B bands, respectively. , , These represent the grayscale values ​​of the red, green, and blue bands of the RGB image before correction. , , These are the grayscale values ​​of the red, green, and blue bands after RGB image correction.

2. The vegetation growth monitoring method according to claim 1, characterized in that, In S3, the Normalized Difference Vegetation Index (NDVI) image is calculated as follows: using the reflectance of the corrected red band image... Reflectance of near-infrared band images The normalized vegetation index (NDVI) image of the overlapping region was calculated: 。 3. A vegetation growth monitoring system for implementing the vegetation growth monitoring method according to any one of claims 1-2, characterized in that, The system includes: A combination camera system placed at a high position, the combination camera system including a visible light monitoring camera and a near-infrared camera; An 18-degree standard gray card is installed on the ground in the monitored area. The 18-degree standard gray card can be captured by the combined camera. The 18-degree standard gray card provides a reference for correction of each band of RGB image and near-infrared image. The data acquisition and transmission device acquires RGB images and near-infrared band images of the monitoring area captured by the combined camera, and transmits the acquired data to a remote monitoring computer.

4. The vegetation growth monitoring system according to claim 3, characterized in that, It also includes a monitoring pole for mounting the combined camera at a high position; and the monitoring pole is also equipped with a solar cell module, which absorbs solar energy and converts it into electrical energy to power the combined camera and the data acquisition and transmission device.

5. The vegetation growth monitoring system according to claim 4, characterized in that, The data acquisition and transmission device includes a rechargeable battery, a power management module, a data acquisition and processing module, and a communication antenna; the rechargeable battery is connected to the solar cell module for storing electrical energy; the data acquisition and processing module transmits the acquired data to a remote monitoring computer through the communication antenna.

6. The vegetation growth monitoring system according to claim 3, characterized in that, In the combined camera setup, the effective pixels of the visible light monitoring camera and the near-infrared camera are: .

7. The vegetation growth monitoring system according to claim 3, characterized in that, The surface of the 18-degree standard gray card is parallel to the camera mirrors of the visible light monitoring camera and the near-infrared camera.

Images