Spectral index construction method and device for detecting tea leaves

Abstract

The invention provides a spectral index construction method and a spectral index construction device for detecting tea leaves. The spectral index construction method comprises the following steps: acquiring multi-spectral reflectivity data of a tea leaf canopy scale; constructing a multispectral-based full-channel difference vegetation index by using multispectral reflectivity data of a tea canopy scale; and constructing a spectral index for detecting the tea based on the full-channel difference vegetation index. By means of the obtained multispectral reflectivity data, the multispectral-based full-channel difference vegetation index is constructed, and physiological parameter changes of different vertical structure positions of the tea leaf canopy can be more sensitively reflected. The index improves the fineness and the accuracy of detection, and is beneficial to finding subtle differences in the growth of the tea leaves. The spectral index for detecting the tea leaves is constructed on the basis of the full-channel difference vegetation index, the spectral index can reflect the growth condition and the nutrition state of the tea leaves more accurately, and the detection accuracy is improved.

Description

Technical Field

[0001] The invention relates to the field of remote sensing technology, and in particular to a spectral index construction method and device for detecting tea leaves. Background Art

[0002] Tea is a popular beverage in the world, and its quality and yield are directly related to the nutritional status of fresh tea leaves. Nitrogen, as an important nutrient element indispensable for plant growth, is a key indicator for measuring the nutritional status of plants. During the growth of tea trees, the nitrogen content has a vital impact on the quality and yield of tea leaves. At the same time, chlorophyll, as the factory for photosynthesis of green plants, also plays a vital role in fresh tea leaves, and is directly related to the synthesis of tea quality components and other organic matter. Therefore, accurately grasping the chlorophyll content and nitrogen content information of tea leaves is extremely important for formulating scientific and reasonable tea garden cultivation and management measures and improving tea quality and yield.

[0003] However, traditional methods for determining chlorophyll and nitrogen content, such as physical and chemical testing in the laboratory, are not only cumbersome, time-consuming and labor-intensive, but also costly and difficult to popularize and apply on a large scale. In addition, these methods are usually destructive and will cause irreversible damage to tea samples, and the test results are lagging, making it difficult to reflect the actual growth status of tea leaves in a timely manner.

[0004] In recent years, spectral measurement technology, as a new type of non-destructive testing technology, has attracted widespread attention for its fast and accurate characteristics. Among them, hyperspectral measurement technology has been applied in many fields due to its high precision. However, the high price of hyperspectral equipment limits its possibility of popularization and application. For this reason, multispectral technology, as an improved version of hyperspectral technology, came into being. While greatly reducing the cost of equipment, it guarantees the accuracy of data to the greatest extent, and has gradually become the preferred solution in various fields.

[0005] Although multispectral technology has many advantages, in previous studies, this type of new non-destructive measurement technology is mostly based on point measurement, such as using SPAD instruments and other non-imaging spectrometers, and the measurement efficiency is relatively low. In addition, when performing spectral measurements on a surface scale, the detection of physiological parameters of targets at different vertical structural positions (such as young and strong leaves of tea) is often not detailed and specific enough. Since young leaves and strong leaves differ in chlorophyll content and other aspects, the detection of their physiological parameters should be carried out separately to obtain more accurate results.

[0006] In view of the limitations of traditional detection methods and the advantages and disadvantages of spectral measurement technology, it is particularly important to find a fast, accurate, economical and practical method to detect the nitrogen content and chlorophyll content of tea. With the rapid development of remote sensing technology, the application of spectral non-destructive detection technology to carry out non-destructive and rapid detection of the physiological and biochemical components of tea has become an important part of current tea growth monitoring and nutritional diagnosis. The application of this technology will provide more accurate and timely information support for the formulation of tea garden cultivation and management measures, and is expected to promote the further improvement of tea quality and yield. Summary of the invention

[0007] The present invention provides a method and device for constructing a spectral index for detecting tea leaves, which is used to solve the defect of insufficiently detailed detection of targets at different vertical structural positions in the prior art and improve the accuracy of multi-spectral detection of tea leaves. The technical solution proposed by the present invention is as follows:

[0008] In a first aspect, the present invention provides a method for constructing a spectral index for detecting tea leaves, comprising:

[0009] Obtain multispectral reflectance data at the tea canopy scale;

[0010] Using the multispectral reflectance data at the tea canopy scale, constructing a multispectral full-channel difference vegetation index;

[0011] A spectral index for detecting tea leaves is constructed based on the full-channel difference vegetation index.

[0012] Optionally, the step of obtaining tea canopy-scale spectral reflectance data includes:

[0013] Acquire multispectral remote sensing image data at the scale of tea canopy, wherein the multispectral remote sensing image data includes DN value data of multiple bands, and each band corresponds to a DN value image;

[0014] Obtain the DN value data of the reference white plate and the known reflectivity data corresponding to the reference white plate in each band;

[0015] The multispectral reflectance data of the tea canopy scale are determined by using the DN value data of the tea canopy scale in the same band, the DN value data of the reference white board and the reflectance data of the corresponding band of the reference white board.

[0016] Optionally, the spectral reflectance data of each band is calculated by the following formula:

[0017] R i =DN i / (DN w / R w )

[0018] In the formula, R iis the spectral reflectance data of the i-th band, DN i is the DN value data of the tea canopy in the i-th band, DN w is the DN value data of the reference white board, R w is the known reflectivity data of the reference white plate corresponding to the i-th band.

[0019] Optionally, the multispectral reflectance data at the tea canopy scale includes the reflectance of the red edge 1 band, the reflectance of the near infrared band, the reflectance of the red edge 2 band, the reflectance of the green band, the reflectance of the blue band and the reflectance of the red band;

[0020] The method of using the multispectral reflectance data at the tea canopy scale to construct a multispectral full-channel difference vegetation index comprises:

[0021] According to the reflectivity of the red edge 1 band, the reflectivity of the near-infrared band, the reflectivity of the red edge 2 band, the reflectivity of the green band, the reflectivity of the blue band and the reflectivity of the red band, the following multispectral full-channel difference vegetation index is constructed:

[0022] FCDVI=2*(ED1+NIR)+ED2+G-8*(B+R)

[0023] Wherein, FCDVI is the full channel difference vegetation index, ED1 is the reflectance of the red edge 1 band, NIR is the reflectance of the near infrared band, ED2 is the reflectance of the red edge 2 band, G is the reflectance of the green band, B is the reflectance of the blue band, and R is the reflectance of the red band.

[0024] Optionally, constructing a spectral index for detecting tea leaves based on the full-channel difference vegetation index includes:

[0025] Obtain a normalized difference vegetation index, and construct a spectral index for detecting tea leaves based on the full-channel difference vegetation index and the normalized difference vegetation index as follows:

[0026] T1=(NDVI-FCDVI) / (NDVI+FCDVI)

[0027] Wherein, TI is the spectral index used to detect tea leaves, NDVI is the normalized difference vegetation index, and FCDVI is the full channel difference vegetation index.

[0028] Optionally, the method further comprises:

[0029] Based on the multispectral reflectance data of the tea canopy scale, a plurality of indices are constructed; wherein the plurality of indices include normalized difference vegetation index, red edge normalized difference index, red edge chlorophyll index, MERIS terrestrial chlorophyll index, modified chlorophyll absorption ratio index, enhanced vegetation index, transformed chlorophyll absorption reflectance index, first correction ratio, second correction ratio, green band chlorophyll index, green normalized difference vegetation index, plant senescence reflectance index, canopy chlorophyll content index, nonlinear vegetation index, and triangular greenness index;

[0030] Constructing a prediction model according to the multiple indices and the normalized difference vegetation index, and training the prediction model to obtain a trained model;

[0031] The chlorophyll content of the tea leaves is predicted based on the trained model.

[0032] In a second aspect, the present invention further provides a spectral index construction device for detecting tea leaves, comprising the following modules:

[0033] Data acquisition module, used to obtain multispectral reflectance data at the tea canopy scale;

[0034] The first construction module is used to construct a multi-spectral full-channel difference vegetation index based on the multi-spectral reflectance data at the tea canopy scale;

[0035] The second building module is used to build a spectral index for detecting tea leaves based on the full-channel difference vegetation index.

[0036] In a third aspect, the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for constructing a spectral index for detecting tea leaves as described in the first aspect above is implemented.

[0037] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for constructing a spectral index for detecting tea leaves as described in the first aspect above.

[0038] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method for constructing a spectral index for detecting tea leaves as described in the first aspect above.

[0039] Based on the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows:

[0040] The spectral index construction method and device for detecting tea leaves provided by the present invention, by acquiring multi-spectral reflectance data at the scale of tea leaves canopy, the data contains reflectance information of different bands, which helps to more accurately evaluate the physiological parameters of tea leaves and can more comprehensively reflect the growth status and nutritional status of tea leaves. By using the acquired multi-spectral reflectance data, a full-channel difference vegetation index based on multi-spectrum is constructed, which can more sensitively reflect the changes in physiological parameters at different vertical structural positions of the tea leaves canopy. This index improves the meticulousness and accuracy of detection and helps to discover subtle differences in tea leaves growth. By constructing a spectral index for detecting tea leaves based on the full-channel difference vegetation index, the spectral index can more accurately reflect the growth status and nutritional status of tea leaves and improve the accuracy of detection. The spectral index constructed by this method is designed for specific physiological parameters of tea leaves (such as chlorophyll content, nitrogen content, etc.), and can perform targeted detection on specific physiological parameters of tea leaves.

[0041] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is a schematic diagram of the flow of the spectral index construction method for detecting tea leaves provided by the present invention.

[0045] Figure 2: is a schematic diagram of the spectral image processing process provided by the present invention. Among them, (a) is the original multispectral image after leaf scale preprocessing, (b) is the multispectral image after leaf scale enhancement using FCDVI, (c) is the image after leaf scale binarization using Otsu, (d) is the multispectral image after leaf scale is masked with the binary image, (e) is the original multispectral image after canopy scale preprocessing, (f) is the multispectral image after canopy scale enhancement using FCDVI, (g) is the image after canopy scale binarization using Otsu, (h) is the multispectral image after canopy scale is masked with the binary image.

[0046] Figure 3 It is the registration fusion effect diagram provided by the present invention. Among them, (a) is the RGB three-channel combined image before the leaf scale multispectral image registration, (b) the leaf scale multispectral image is the RGB three-channel combined true color image after the registration, (c) is the RGB three-channel combined image before the canopy scale multispectral image registration, and (d) is the RGB three-channel combined true color image after the canopy scale multispectral image registration.

[0047] Figure 4 It is a schematic diagram of the structure of a spectral index construction device for detecting tea leaves provided by the present invention.

[0048] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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] Combine the following Figure 1-Figure 4 The present invention describes the method and device for constructing a spectral index for detecting tea leaves.

[0051] Reference Figure 1 As shown, the spectral index construction method for detecting tea leaves includes the following:

[0052] Step S110, obtaining multispectral reflectance data at the tea canopy scale.

[0053] The goal of this step is to collect reflectance data of the tea canopy in different spectral bands. These data will be used for subsequent spectral index construction and analysis. Multispectral remote sensing instruments, such as multispectral cameras or spectrometers, can be used, which are capable of acquiring reflectance data in multiple bands simultaneously. Make sure that the weather is clear and cloud-free during the measurement to reduce the impact of atmospheric conditions on the measurement results. At the same time, the measurement should be performed at a certain height above the tea canopy to ensure that the acquired data can represent the average reflectance of the entire canopy. Record the reflectance data of each band and save it in digital format for subsequent processing and analysis.

[0054] Step S120: constructing a multispectral full-channel difference vegetation index based on multispectral reflectance data at the tea canopy scale.

[0055] The full-channel difference vegetation index that can reflect the growth status and nutritional status of tea leaves is constructed using the acquired multispectral reflectance data. From the acquired multispectral data, bands that are sensitive to tea growth and can reflect its nutritional status are selected. These bands can include red edge 1 band (ED1), near infrared band (NIR), red edge 2 band (ED2), green band (G), blue band (B) and red band (R), etc., each of which plays a vital role in vegetation monitoring. Through the spectral reflectance data of these bands, the full-channel difference vegetation index is constructed by adding, subtracting or weighted summing. This index integrates the information of multiple bands and can comprehensively reflect the differences in the reflectance characteristics of the tea canopy in different bands, avoiding the information loss that may be caused by a single band, thereby more comprehensively reflecting the reflectance characteristics of the tea canopy and its relationship with the growth status. The calculated FCDVI value can be used as an important indicator for tea growth monitoring to evaluate the growth status, nutritional level and possible growth stress of tea leaves.

[0056] Step S130: constructing a spectral index for detecting tea leaves based on the full-channel difference vegetation index.

[0057] In the process of tea garden management and tea quality assessment, accurate spectral index is the key to measure the growth status and nutritional level of tea. In step S130, based on the full channel difference vegetation index, a spectral index specifically used for detecting tea is further constructed. The construction of this index can be based on the above-mentioned full channel difference vegetation index FCDVI, as well as other vegetation indices, such as normalized difference vegetation index (NDVI), chlorophyll vegetation index (Chlorophyll Vegetation Index, CVI), soil-adjusted vegetation index (Soil-Adjusted Vegetation Index, SAVI), enhanced vegetation index (Enhanced Vegetation Index, EVI), red edge position index (Red Edge Position Index, REPI), normalized water index (Normalized Water Index, NWI), etc.

[0058] The above CVI uses the specific absorption characteristics of chlorophyll in the red and near-infrared bands to evaluate the chlorophyll content by calculating the reflectivity difference between these two bands. For tea, CVI can reflect the relative content of chlorophyll in tea leaves, and thus indirectly indicate the growth status and nutritional status of tea. The above SAVI corrects the soil background effect on the basis of NDVI, and is particularly suitable for environments with large soil background changes. In tea-growing areas, if the soil background has a significant impact on spectral reflectance, SAVI can provide more accurate vegetation information. The above EVI is an improvement on NDVI. By introducing atmospheric correction factors and soil adjustment coefficients, the assessment accuracy of vegetation coverage and biomass is improved. In tea garden management, EVI can more effectively monitor the growth dynamics and health of tea. The red edge position refers to the steep edge position of the transition from the red light region to the near-infrared region in the chlorophyll absorption spectrum. The above REPI evaluates the chlorophyll content and plant health by measuring the changes in this position. For tea, REPI can reflect subtle changes in chlorophyll in tea leaves, which helps to detect growth abnormalities in a timely manner. The above NWI is mainly used to monitor water bodies and soil moisture, but in tea garden management, it can also be used as an auxiliary indicator to assess the moisture status of tea leaves and irrigation needs. By monitoring the moisture content of tea leaves, NWI helps guide the moisture management of tea gardens.

[0059] These indices are based on the reflectance characteristics of tea leaves in different spectral bands and can reflect the growth status and nutritional level of tea leaves. Since the application scenarios and sensitivities of different spectral indices may vary, technicians in this field can comprehensively consider the specific monitoring needs and tea garden environment and select one or more of the above indices to construct a spectral index for detecting tea leaves.

[0060] The spectral index construction method for detecting tea leaves provided by the present invention obtains multi-spectral reflectance data at the scale of tea canopy, and the data contains reflectance information of different bands, which is helpful to more accurately evaluate the physiological parameters of tea leaves and can more comprehensively reflect the growth status and nutritional status of tea leaves. By using the acquired multi-spectral reflectance data, a full-channel difference vegetation index based on multi-spectrum is constructed, which can more sensitively reflect the changes in physiological parameters of different vertical structural positions of tea canopy. This index improves the meticulousness and accuracy of detection and helps to find subtle differences in tea growth. By constructing a spectral index for detecting tea leaves based on the full-channel difference vegetation index, the spectral index can more accurately reflect the growth status and nutritional status of tea leaves and improve the accuracy of detection. The spectral index constructed by the method is designed for specific physiological parameters of tea leaves (such as chlorophyll content, nitrogen content, etc.), and can perform targeted detection on specific physiological parameters of tea leaves.

[0061] The spectral index construction method for detecting tea leaves provided by the present invention first obtains multispectral reflectance data at the tea canopy scale through remote sensing technology or a dedicated spectrometer. This process is relatively simple and fast, avoiding the cumbersome sampling, processing and testing steps in traditional laboratory physical and chemical testing, thereby greatly saving time and labor costs. After obtaining the multispectral reflectance data, the method uses a computer for automated processing and analysis, further improving the detection efficiency.

[0062] The spectral index construction method for detecting tea leaves provided by the present invention can simultaneously obtain the reflectance data of multiple points of the tea canopy by using a multi-spectral remote sensing instrument and a surface-scale measurement method, thereby avoiding the low efficiency problem of point measurement. Through surface-scale measurement, the method can more comprehensively evaluate the growth status and nutritional status of tea leaves, and provide more accurate information support for the formulation of tea garden cultivation management measures. The method adopts spectral non-destructive detection technology, does not cause any damage to tea samples, and realizes real-time monitoring of tea growth conditions. By quickly acquiring and processing multi-spectral reflectance data, the method can timely reflect the growth status and nutritional status of tea leaves, and provide timely decision-making support for tea garden management.

[0063] In an optional embodiment, the step S110 of obtaining tea canopy-scale spectral reflectance data includes:

[0064] S1101. Acquire multispectral remote sensing image data at the scale of tea canopy, wherein the multispectral remote sensing image data includes DN value data of multiple bands, and each band corresponds to a DN value image.

[0065] This step is to obtain remote sensing image data of tea canopy in different spectral bands. These data are usually expressed in digital form, namely DN value (Digital Number) data. Multispectral remote sensing instruments can capture images of multiple bands at the same time, and each band corresponds to a DN value image. These bands cover visible light, near infrared and other regions, and can reflect the reflectance characteristics of tea canopy in different spectral ranges.

[0066] S1102, obtaining DN value data of a reference white plate and known reflectivity data corresponding to each waveband of the reference white plate.

[0067] The reference white board is a standard object used to calibrate remote sensing instruments, and its reflectivity is known in each band. Before or after obtaining multispectral remote sensing image data at the tea canopy scale, it is necessary to measure the DN value data of the reference white board. At the same time, it is also necessary to obtain the known reflectivity data corresponding to the reference white board in each band, which is provided by the manufacturer or obtained through strict calibration.

[0068] S1103, using the DN value data of the tea canopy scale in the same band, the DN value data of the reference white board, and the reflectance data of the corresponding band of the reference white board, determine the multispectral reflectance data of the tea canopy scale.

[0069] For each band, the reflectance of the tea canopy can be calculated using the following formula:

[0070] Tea canopy reflectivity = (tea canopy DN value - dark current DN value) / (reference white board DN value - dark current DN value) × reference white board reflectivity

[0071] The dark current DN value refers to the DN value output by the instrument under no light conditions, which is used to calibrate and remove noise inside the instrument. In actual operation, since the dark current DN value is usually small and stable, it can be ignored to simplify the calculation process. Therefore, the above formula can be simplified as follows:

[0072] R i =DN i / (DN w / R w )

[0073] In the formula, R i is the spectral reflectance data of the i-th band, DN i is the DN value data of the tea canopy in the i-th band, DN w is the DN value data of the reference white board, R w is the known reflectivity data of the reference white plate corresponding to the i-th band (ie, the standard reflectivity of the reference white plate in this band).

[0074] Through this step, the reflectance data of the tea canopy in each band can be obtained, which will be used for subsequent spectral index construction and analysis.

[0075] Remote sensing instruments may be affected by internal noise, optical distortion and other factors when collecting data, resulting in data errors. When collecting data, external environmental factors (such as light intensity, atmospheric conditions, etc.) may also affect the data. The present invention calibrates by reference to a whiteboard, which can eliminate the influence of internal instrument noise and external environmental factors on the data, thereby improving the accuracy of tea canopy scale spectral reflectance data. Using the same reference whiteboard for calibration can ensure that data collected at different times and locations are consistent and comparable, which is convenient for subsequent data analysis and application.

[0076] In an optional embodiment, the present invention uses a MS600 V2 multispectral camera to collect multispectral remote sensing image data. The camera has multiple channels, including Blue (B), Green (G), Red (R), Red Edge (ED1), Red EdgeLP (ED2) and Near Infrared (NIR). Since the sensors of each channel are distributed in an array, the acquired images of each channel may have deviations in spatial position. However, the camera is not equipped with an automatic registration program. In order to facilitate band fusion and spectral information sampling, the multispectral remote sensing image data is a multispectral image. First, the average DN value of the whiteboard in each image is extracted from the multispectral image, and the multispectral DN value image is converted into a reflectance image in python3.6 according to formula 1 in combination with the reflectance of each band calibrated by the reference whiteboard. Then, the Sift algorithm is used to automatically select and match the features of each band image, and then the band r fusion is performed.

[0077] Specifically, the multispectral image is preprocessed first. This includes: in the collected multispectral images, finding and extracting the average DN (Digital Number) value of the white board in each image. The DN value is the brightness value of the image pixel, which is used to represent the brightness information of the image. The white board is used for calibration and standardization, because its reflectivity is close to 100%, so it can be used as a reference for reflectivity calculation. Combined with the reflectivity of each band calibrated by the reference white board, the multispectral DN value image is converted into a reflectivity image using Formula 1. The converted reflectivity image is more convenient for subsequent spectral information analysis and processing.

[0078] In order to solve the problem of deviation in spatial position of each channel image, Sift (Scale-Invariant Feature Transform) algorithm is used for image registration. Sift algorithm is an algorithm for detecting and describing local features of images. It can identify stable feature points under different scales, rotations, illumination and other conditions. Sift algorithm is used to automatically select feature points in each band image and perform feature matching. Feature matching refers to finding the same feature points in different images to determine the relative position relationship between images. Based on feature matching, band fusion is performed. Band fusion refers to merging images of multiple bands into one image according to certain rules for subsequent spectral information sampling and analysis. For example, the pixel values ​​of each channel image are aligned and combined according to their spatial positions. Finally, the effects before and after registration and fusion are shown. By comparing the images before and after registration, the effects of image registration and fusion can be intuitively seen, as well as whether the consistency of the spatial position of each channel image is improved.

[0079] The effect before and after registration fusion is as follows Figure 3 As shown. Among them, (a) is the RGB three-channel combined image before the leaf-scale multispectral image is registered, (b) the leaf-scale multispectral image is the RGB three-channel combined true color image after the registration, (c) is the RGB three-channel combined image before the canopy-scale multispectral image is registered, and (d) is the RGB three-channel combined true color image after the canopy-scale multispectral image is registered.

[0080] Before fusion, the images of each spectral band may have problems such as spatial deviation, uneven brightness or noise pollution due to factors such as sensor array distribution, atmospheric interference, instrument noise, etc. Through fusion calibration, the spatial deviation of the images of each band is corrected, the brightness is uniformly adjusted, the noise is suppressed, and the overall image quality is significantly improved. The image is clearer and more delicate, and easier to analyze and process later. In addition, before fusion, due to factors such as sensor array distribution and internal camera processing, the images of each band may have deviations in spatial position, resulting in misalignment or overlap when images are stitched. Through feature matching and band fusion technology, the spatial positions of the images of each band are accurately aligned, and the image stitching is more natural and accurate. The spatial consistency is significantly improved, providing a reliable foundation for subsequent spectral information sampling and analysis.

[0081] In an optional embodiment, in order to effectively extract the discrete spectral curve of the canopy part and avoid the influence of background such as shadows in the image, the present invention proposes a full channel difference vegetation index (FCDVI) based on multispectral images by performing spectral feature analysis and multiple tests on the canopy and other objects. The multispectral reflectance data of the tea canopy scale includes the reflectance of the red edge 1 band, the reflectance of the near infrared band, the reflectance of the red edge 2 band, the reflectance of the green band, the reflectance of the blue band and the reflectance of the red band, as shown in Table 1.

[0082]

[0083] Table 1

[0084] The step S120 described above uses the multispectral reflectance data at the tea canopy scale to construct a multispectral full-channel difference vegetation index, including:

[0085] S1201. Construct a multispectral full-channel difference vegetation index according to the reflectivity of the red edge 1 band, the reflectivity of the near infrared band, the reflectivity of the red edge 2 band, the reflectivity of the green band, the reflectivity of the blue band, and the reflectivity of the red band:

[0086] FCDVI=2*(ED1+NIR)+ED2+G-8*(B+R), formula 2;

[0087] In the formula, FCDVI is the full channel difference vegetation index, ED1 is the reflectance of the red edge 1 band, NIR is the reflectance of the near infrared band, ED2 is the reflectance of the red edge 2 band, G is the reflectance of the green band, B is the reflectance of the blue band, and R is the reflectance of the red band. This index can greatly enhance the characteristics of tea leaves in multispectral images for both leaf-scale and canopy-scale multispectral images, and increase the contrast with other objects to facilitate the distinction between tea leaves and background.

[0088] The full-channel difference vegetation index FCDVI of the present invention combines the reflectance data of multiple spectral bands, and can more comprehensively reflect the vegetation characteristics of the tea canopy. Compared with the vegetation index of a single spectral band, FCDVI has higher precision and accuracy in vegetation monitoring. Through a specific mathematical model, FCDVI can enhance the vegetation information of the tea canopy and weaken the influence of background information. This helps to accurately identify the tea canopy in a complex environment and provide reliable data support for subsequent remote sensing applications. The construction process of FCDVI takes into account the data of multiple spectral bands, so it has strong adaptability. It can be used under different lighting conditions, observation angles and surface cover types, providing a wider range of application scenarios for remote sensing monitoring of tea canopies.

[0089] like Figure 2 The figure shows the schematic diagram of the spectral image processing process. Among them, (a) is the original multispectral image after leaf scale preprocessing, (b) is the multispectral image after leaf scale enhancement using FCDVI, (c) is the image after leaf scale binarization using Otsu, (d) is the multispectral image after leaf scale using the binary image to mask the original multispectral image, (e) is the original multispectral image after canopy scale preprocessing, (f) is the multispectral image after canopy scale enhancement using FCDVI, (g) is the image after canopy scale binarization using Otsu, and (h) is the multispectral image after canopy scale using the binary image to mask the original multispectral image.

[0090] First of all, Figure 2 (a) to (d) in the figure are analyzed: (a) is the original multispectral image after leaf scale preprocessing, which contains noise, background interference, and reflectance differences between different leaves. The leaf scale is enhanced using FCDVI. The multispectral image (b) reduces noise and background interference, making the leaf features more prominent. The leaf scale is converted into a binary image containing only black and white colors after Otsu binarization (c). The leaves are identified as white (foreground), while the background and other non-leaf parts are identified as black (background). The binary image simplifies image analysis. The binary image is used as a mask to extract the leaf area in the original multispectral image, and the multispectral image (d) is obtained after the leaf scale is masked with the binary image of the original multispectral image. The masked image only contains the information of the leaf area and removes the interference of the background and other non-leaf parts, which helps to more accurately analyze the spectral characteristics of the leaves.

[0091] Secondly, Figure 2 (e) to (h) are analyzed: Similar to the original image at the leaf scale, (e) is the original multispectral image after canopy scale preprocessing. It contains the reflectance or transmittance information of the entire canopy in different spectral bands. Similar to the leaf scale, FCDVI is used to enhance the vegetation characteristics at the canopy scale, and the multispectral image enhanced by FCDVI at the canopy scale is obtained (f). The enhanced image may more clearly show the chlorophyll content, health status, canopy structure and other characteristics of the canopy. At the canopy scale, Otsu binarization is used to convert the image into a binary image to distinguish the canopy from the background, and the canopy scale image after Otsu binarization is obtained (g). At the canopy scale, the binary image is used as a mask to extract the canopy area in the original multispectral image, and the canopy scale is masked with the binary image to obtain the multispectral image (h). The masked image only contains the information of the canopy area, removing the interference of the background and other non-canopy parts. This helps to more accurately analyze the spectral characteristics of the canopy and reduce the impact of background noise.

[0092] At both the leaf scale and canopy scale, the multispectral images enhanced with FCDVI show vegetation features more clearly and help reduce noise and background interference. Otsu binarization converts images into binary images, simplifying image analysis. Using the binary image as a mask can extract the leaf or canopy area of ​​interest, allowing for more accurate analysis of spectral features. The masked multispectral image removes interference from the background and other non-interested areas, making subsequent analysis more accurate and reliable, and helping to improve the accuracy and efficiency of image analysis.

[0093] The original multispectral image and two groups of canopy-scale multispectral images enhanced by FCDVI are used, and the spectral index calculation formula for tea detection of the present invention (i.e., Formula 3) is applied to the enhanced multispectral images. The calculated vegetation index is paired with the physiological parameters of tea leaves (such as nitrogen content, chlorophyll content, etc.) to form a data set. Statistical methods (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) are used to evaluate the correlation between vegetation index and physiological parameters. The correlation between vegetation index and physiological parameters obtained by comparing the original image and the image enhanced by FCDVI is analyzed, and the specific effect of FCDVI processing on improving the correlation is analyzed.

[0094] Through comparative analysis, it is found that the multispectral images processed by FCDVI show significant advantages in the correlation between vegetation index and tea physiological parameters, which is specifically reflected in the increase in the number of significantly correlated data pairs and the improvement in correlation coefficients. This proves that the FCDVI for the MS600 V2 multispectral sensor proposed in the present invention can significantly enhance the characteristics of tea leaves and is of great help in removing soil and shadow noise.

[0095] In an optional embodiment, in order to more accurately detect the growth status and health of tea leaves, the present invention constructs a new spectral index (TI) for detecting tea leaves based on the full channel difference vegetation index (FCDVI) and the normalized difference vegetation index (NDVI). This index combines the advantages of the two vegetation indices and aims to improve the accuracy and reliability of tea leaf detection. The spectral index for detecting tea leaves based on the full channel difference vegetation index described in step S130 includes:

[0096] S1301, obtaining a normalized difference vegetation index, and constructing a spectral index for detecting tea leaves based on the full-channel difference vegetation index and the normalized difference vegetation index as follows:

[0097] T1=(NDVI-FCDVI) / (NDVI+FCDVI), formula 3;

[0098] Wherein, TI is the spectral index used to detect tea leaves, NDVI is the normalized difference vegetation index, and FCDVI is the full channel difference vegetation index.

[0099] The Normalized Difference Vegetation Index (NDVI) is one of the important parameters that reflect the growth and nutritional information of crops. Its calculation formula is NDVI = (NIR-R) / (NIR+R), where NIR is the reflectance of the near-infrared band and R is the reflectance of the red band. The reflectance data of the near-infrared band and the red band of the tea canopy are obtained through remote sensing images or ground-based spectral measurement equipment, and then substituted into the NDVI formula for calculation to obtain the NDVI value.

[0100] The full channel difference vegetation index (FCDVI) is obtained by combining the reflectance data of multiple spectral bands of the tea canopy through a mathematical model such as formula 2. Similarly, the reflectance data of these bands are obtained through remote sensing images or ground spectral measurement equipment, and then substituted into the FCDVI formula for calculation to obtain the FCDVI value.

[0101] Based on the obtained NDVI and FCDVI values, a spectral index (TI) for detecting tea leaves was constructed. Its calculation formula is the above formula 3. Through this formula, the difference and sum of NDVI and FCDVI can be combined to form a new spectral index TI, which is used to reflect the specific vegetation characteristics of the tea canopy.

[0102] NDVI and FCDVI have different emphases and sensitivities in reflecting vegetation characteristics. NDVI mainly reflects the coverage and growth status of vegetation, while FCDVI combines information from multiple spectral bands and can reflect the vegetation characteristics of the tea canopy in more detail. Combining the two can complement each other and more comprehensively reflect the growth status and health of tea. By constructing the spectral index TI, the advantages of NDVI and FCDVI can be combined. TI can more accurately reflect the vegetation characteristics of the tea canopy and improve the accuracy and reliability of tea detection. In tea remote sensing monitoring, TI can more accurately identify the tea canopy and extract key information related to tea growth. This helps to reduce misjudgments and missed judgments and improve the accuracy of tea identification.

[0103] By combining the advantages of the two vegetation indices NDVI and FCDVI, the TI can more comprehensively reflect the vegetation characteristics of the tea canopy, thereby improving the accuracy of tea detection. NDVI is sensitive to vegetation coverage and growth conditions, while FCDVI combines information from multiple spectral bands, which can enhance vegetation information and weaken the influence of background information. Therefore, TI can more accurately reflect the growth status and health status of tea. The construction process of TI takes into account data from multiple spectral bands, so it has a strong information extraction capability. It can accurately identify the tea canopy in a complex environment and extract key information related to tea growth, providing reliable data support for subsequent remote sensing applications. By monitoring the TI changes of the tea canopy, information such as the growth status, health status and yield potential of the tea can be timely understood. This helps to guide the production and management of tea, improve the quality and yield of tea, and provide strong support for the sustainable development of the tea industry.

[0104] The method of the present invention ensures that the selected bands can sensitively reflect the growth status and nutritional status of tea leaves by selecting and optimizing the bands of multispectral data. These bands generally cover regions such as visible light, near infrared and short-wave infrared, and can capture the reflective characteristics of tea canopies in different spectral ranges. After obtaining the multispectral reflectance data, the method performs data preprocessing, including removing noise, correcting instrument errors, etc., to ensure the accuracy and reliability of the data. This step helps to reduce errors caused by instruments or environmental factors, and provides a high-quality data basis for subsequent analysis. The method adopts a canopy scale measurement method to obtain multispectral reflectance data of the entire tea canopy through remote sensing technology or a special spectrometer. This method can reduce the interference of the background environment (such as soil, weeds, etc.) on the extraction of tea information, because the canopy reflectance mainly reflects the growth status of the tea leaves themselves. By constructing a full-channel difference vegetation index based on multispectral, the method can further weaken the impact of the background environment. The vegetation index is a comprehensive indicator reflecting the growth status of vegetation. By calculating the difference in reflectance of different bands, the characteristic information of the tea canopy can be highlighted and the interference of the background environment can be reduced. The spectral index constructed by the method of the present invention is designed for specific physiological parameters of tea leaves (such as chlorophyll content, nitrogen content, etc.). This targeted design enables the spectral index to more accurately reflect the growth status and nutritional status of tea leaves, thereby improving the accuracy of detection.

[0105] In an optional embodiment, multiple vegetation indices can be constructed based on multispectral reflectance data, and these indices can be used together with the normalized difference vegetation index (NDVI) to construct a prediction model to predict the chlorophyll content of tea leaves. The present invention selects multiple vegetation indices and original bands as variables, which are used for subsequent model training. Although the vegetation index has a low dimension, the selected machine learning algorithm can handle redundancy problems such as collinearity between variables, so all selected variables are included in the model.

[0106] The method further comprises:

[0107] S210. Based on the multispectral reflectance data of the tea canopy scale, multiple indices are constructed; wherein the multiple indices include normalized difference vegetation index, red edge normalized difference index, red edge chlorophyll index, MERIS terrestrial chlorophyll index, modified chlorophyll absorption ratio index, enhanced vegetation index, transformed chlorophyll absorption reflectance index, first correction ratio, second correction ratio, green band chlorophyll index, green normalized difference vegetation index, plant aging reflectance index, canopy chlorophyll content index, nonlinear vegetation index, triangular greenness index. The formulas of these indices are shown in Table 2. ED in Table 2 refers to the reflectance of the red edge band, and NIR2 refers to the reflectance of the near infrared second band. The meanings of other parameters are consistent with those described above.

[0108]

[0109] Table 2

[0110] Collect multispectral reflectance data at the tea canopy scale. These data are usually obtained through remote sensing equipment or ground-based spectral measurement equipment, covering information in different spectral bands. Based on the collected multispectral reflectance data, construct multiple vegetation indices. These indices include Normalized Difference Vegetation Index (NDVI), Normalized Difference Red Edge (NDRE), Chlorophyll Index Red Edge (CIRE), MERIS Terrestrial Chlorophyll Index (MTCI), Modified Chlorophyll Absorption Ratio Index (MCARI), Enhanced Vegetation Index (EVI), Transformed Chlorophyll Absorption Reflectance Index (TCARI), Modified Simple Ratio 1 (MSR1), Modified Simple Ratio 2 (MSR2), Chlorophyll Index Green (CIG), Green Normalized Difference Vegetation Index (GNDVI), Plant Senescence Reflectance Index (Plant Senescence Reflectance Index, GNDVI), and other indices. The indexes include the Medium Resolution Imaging Spectroradiometer (MERIS) and the Chlorophyll Absorption Reflectance Index (CARI). These indices can reflect the vegetation characteristics of the tea canopy, such as chlorophyll content, vegetation coverage, and growth status.

[0111] S220: construct a prediction model according to the multiple indices and the normalized difference vegetation index, train the prediction model, and obtain a trained model.

[0112] The constructed multiple vegetation indices and NDVI are used as input variables to jointly construct a prediction model. Appropriate machine learning algorithms, such as support vector machine (SVM), partial least squares (PLS), neural network, etc., are selected to train the model. Through the training process, the parameters and structure of the model are optimized so that it can accurately predict the chlorophyll content of tea leaves.

[0113] The above-mentioned random forest (RF) is an ensemble learning method that improves the accuracy and stability of the model by constructing multiple decision trees and synthesizing their prediction results. Each decision tree is trained based on randomly selected bootstrap samples and input variables. Partial least squares regression (PLS) is a multivariate statistical data analysis method that is particularly suitable for dealing with situations where there is multicollinearity between variables or the number of variables is greater than the number of samples. It extracts the dominant factors with the strongest explanatory power for the dependent variable by combining the advantages of principal component analysis, canonical analysis, and linear regression. Support vector machine (SVM) is a machine learning algorithm based on kernel functions that can map input variables to high-dimensional feature space to process high-dimensional input vectors. SVM has been widely used in fields such as spectral analysis and produces accurate calibration results. BP neural network is a multilayer feedforward neural network that is trained using an error back propagation algorithm. It contains an input layer, a hidden layer, and an output layer, and minimizes the prediction error by adjusting network parameters.

[0114] S230, predicting the chlorophyll content of tea leaves based on the trained model.

[0115] The trained model is validated using an independent validation dataset to evaluate its prediction performance and accuracy. If the model performs well, it can be applied to actual tea production to predict the chlorophyll content of tea and guide tea planting and management.

[0116] The present invention can make full use of the information in the multispectral reflectance data and improve the accuracy of predicting the chlorophyll content of tea by constructing multiple vegetation indices and combining them with NDVI to jointly construct a prediction model. Different vegetation indices can reflect the different characteristics of the tea canopy. Combining them can form a complementary relationship and improve the prediction ability of the model. Using multiple vegetation indices as input variables can increase the diversity and complexity of the model, so that it can better adapt to the growth conditions of tea leaves under different environments and conditions. This helps to enhance the robustness of the model so that it can maintain stable prediction performance in tea production at different times and locations. By predicting the chlorophyll content of tea leaves, the growth status and health status of tea leaves can be understood in a timely manner, providing strong support for the production management of tea leaves. The method makes full use of the advantages of remote sensing technology, constructs multiple vegetation indices through multispectral reflectance data, and is successfully applied to the prediction of chlorophyll content of tea leaves.

[0117] In order to evaluate the performance of the model, the present invention adopts the leave-one-out cross-validation method. This method divides the data set into n subsets, leaving one subset as the test set each time, and the remaining subsets are used as training sets for model training. Then, the trained model is used to predict the test set and the accuracy of the prediction results is calculated. This process is repeated n times, and each subset is used as a test set.

[0118] In order to quantify the accuracy of the model, the present invention calculates two indicators, goodness of fit R2 and root mean square error RMSE. The calculation formula can refer to the description in the prior art. The closer R2 is to 1, the better the model fit is; the smaller RMSE is, the smaller the error of the prediction result is and the higher the accuracy is.

[0119] The present invention constructs a prediction model by selecting a variety of machine learning algorithms and vegetation indices, and verifies the performance of the model by precision evaluation methods such as leave-one-out cross validation, R2 and RMSE. This method provides an effective solution for vegetation index and modeling, and helps to improve the accuracy and stability of the model. Those skilled in the art can also select multiple indices from the above to construct a prediction model for nitrogen content or other factors related to tea growth.

[0120] The spectral index construction device for detecting tea leaves provided by the present invention is described below. The spectral index construction device for detecting tea leaves described below and the spectral index construction method for detecting tea leaves described above can be referred to each other.

[0121] The spectral index construction device for detecting tea leaves provided by the present invention refers to Figure 4 As shown, including:

[0122] The data acquisition module 310 is used to acquire multispectral reflectance data at the tea canopy scale;

[0123] The first construction module 320 is used to construct a multi-spectral full-channel difference vegetation index based on the multi-spectral reflectance data at the tea canopy scale;

[0124] The second construction module 330 is used to construct a spectral index for detecting tea leaves based on the full-channel difference vegetation index.

[0125] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the spectral index construction method for detecting tea leaves.

[0126] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0127] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the spectral index construction method for detecting tea leaves provided by the above-mentioned methods.

[0128] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the spectral index construction method for detecting tea leaves provided by the above-mentioned methods.

[0129] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a spectral index for detecting tea leaves, characterized in that: include: Obtain multispectral reflectance data at the tea canopy scale; Using the multispectral reflectance data at the tea canopy scale, constructing a multispectral full-channel difference vegetation index; A spectral index for detecting tea leaves is constructed based on the full-channel difference vegetation index.

2. The method for constructing a spectral index for detecting tea leaves according to claim 1, characterized in that: The method of obtaining tea canopy-scale spectral reflectance data includes: Acquire multispectral remote sensing image data at the scale of tea canopy, wherein the multispectral remote sensing image data includes DN value data of multiple bands, and each band corresponds to a DN value image; Obtain the DN value data of the reference white plate and the known reflectivity data corresponding to the reference white plate in each band; The multispectral reflectance data of the tea canopy scale are determined by using the DN value data of the tea canopy scale in the same band, the DN value data of the reference white board and the reflectance data of the corresponding band of the reference white board.

3. The method for constructing a spectral index for detecting tea leaves according to claim 2, characterized in that: The spectral reflectance data of each band is calculated by the following formula: R i =DN i / (DN w / R w ) In the formula, R i is the spectral reflectance data of the i-th band, DN i is the DN value data of the tea canopy in the i-th band, DN w is the DN value data of the reference white board, R w is the known reflectivity data of the reference white plate corresponding to the i-th band.

4. The method for constructing a spectral index for detecting tea leaves according to claim 1, characterized in that: The multispectral reflectance data at the tea canopy scale include the reflectance of the red edge 1 band, the reflectance of the near infrared band, the reflectance of the red edge 2 band, the reflectance of the green band, the reflectance of the blue band and the reflectance of the red band; The method of using the multispectral reflectance data at the tea canopy scale to construct a multispectral full-channel difference vegetation index comprises: According to the reflectivity of the red edge 1 band, the reflectivity of the near-infrared band, the reflectivity of the red edge 2 band, the reflectivity of the green band, the reflectivity of the blue band and the reflectivity of the red band, the following multispectral full-channel difference vegetation index is constructed: FCDVI=2*(ED1+NIR)+ED2+G-8*(B+R) Where, FCDVI is the full channel difference vegetation index, is the reflectivity of the red edge 1 band, NIR is the reflectivity of the near infrared band, ED2 is the reflectivity of the red edge 2 band, G is the reflectivity of the green band, B is the reflectivity of the blue band, and R is the reflectivity of the red band.

5. The method for constructing a spectral index for detecting tea leaves according to claim 1, characterized in that: The method of constructing a spectral index for detecting tea leaves based on the full-channel difference vegetation index comprises: Obtain a normalized difference vegetation index, and construct a spectral index for detecting tea leaves based on the full-channel difference vegetation index and the normalized difference vegetation index as follows: T1=(NDVI-FCDVI) / (NDVI+FCDVI) Wherein, TI is the spectral index used to detect tea leaves, NDVI is the normalized difference vegetation index, and FCDVI is the full channel difference vegetation index.

6. The method for constructing a spectral index for detecting tea leaves according to claim 1, characterized in that: The method further comprises: Based on the multispectral reflectance data of the tea canopy scale, a plurality of indices are constructed; wherein the plurality of indices include normalized difference vegetation index, red edge normalized difference index, red edge chlorophyll index, MERIS terrestrial chlorophyll index, modified chlorophyll absorption ratio index, enhanced vegetation index, transformed chlorophyll absorption reflectance index, first correction ratio, second correction ratio, green band chlorophyll index, green normalized difference vegetation index, plant senescence reflectance index, canopy chlorophyll content index, nonlinear vegetation index, and triangular greenness index; Constructing a prediction model according to the multiple indices and the normalized difference vegetation index, and training the prediction model to obtain a trained model; The chlorophyll content of the tea leaves is predicted based on the trained model.

7. A spectral index construction device for detecting tea leaves, characterized in that: include: Data acquisition module, used to obtain multispectral reflectance data at the tea canopy scale; The first construction module is used to construct a multi-spectral full-channel difference vegetation index based on the multi-spectral reflectance data at the tea canopy scale; The second building module is used to build a spectral index for detecting tea leaves based on the full-channel difference vegetation index.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for constructing a spectral index for detecting tea leaves as claimed in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for constructing a spectral index for detecting tea leaves as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for constructing a spectral index for detecting tea leaves as claimed in any one of claims 1 to 6 is implemented.

Images