A method for identifying camouflage materials in northern environments based on spectral characteristic index
By adopting a recognition method based on spectral feature index in the northern environment, the spectral feature differences between the jungle camouflage camouflage net and the background environment vegetation are extracted, and the ratio camouflage index RCI is constructed, which solves the problem of camouflage material recognition in the northern environment and achieves higher recognition accuracy and adaptability.
Patent Information
- Application Number
- CN202210736637.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-27
- Publication Date
- 2025-05-16
- Estimated Expiration
- 2042-06-27
AI Technical Summary
The prior art is difficult to effectively identify camouflage materials in northern environments, especially in complex background environments, resulting in insufficient accuracy of camouflage identification.
Using a recognition method based on the spectral feature index, the VIS-NIR spectral data of jungle camouflage camouflage net samples and typical vegetation in the background environment were obtained, and the identification feature bands were extracted using spectral similarity measurement and continuum removal method, ratio camouflage index RCI was constructed, and identification threshold conditions were set to identify the camouflage targets of hyperspectral images.
It improves the accuracy of camouflage target recognition in the northern environment, is suitable for the recognition of camouflage camouflage networks in the northern environment, and achieves higher adaptability and invariance.
Smart Images
Figure CN115204219B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of remote sensing hyperspectral data processing, and more particularly to a method for identifying camouflage materials in a northern environment based on spectral characteristic indexes. Background Art
[0002] In modern information warfare, the “discovery means destruction” has been gradually realized for battlefield targets, and camouflage technology has become an effective means to interfere with enemy detection and protect combat equipment and personnel. With the rapid development of material technology, target camouflage means and strategies are also constantly improving, especially the use of advanced coatings and camouflage nets, which can achieve the effect of “different objects with the same spectrum” between the target and the background under certain background environments, increasing the difficulty of camouflage identification. Therefore, it is of great significance to carry out research on the analysis of camouflage material characteristics and identification methods.
[0003] Based on traditional visible light or multi-spectral remote sensing technology, it is difficult to identify camouflaged targets due to the wide spectral range covered by a single band, which brings great challenges to camouflage identification. Hyperspectral remote sensing technology has the characteristics of wide spectral range, multiple bands, and high spectral resolution (nanometer level), which provides extremely rich information for understanding ground objects. The difference between the target and the background is the fundamental reason for the exposure of the target. The spectral information of the ground object has a fingerprint effect on target identification. By analyzing the spectral characteristics of the ground object and extracting the characteristic index based on it, which is related to the characteristics of the material, it can provide a basis for fine identification of the ground object. Especially in the detection of surface resources and environment, hyperspectral technology effectively improves the accuracy of ground object identification and classification.
[0004] At present, the application of hyperspectral remote sensing technology in the analysis and identification of camouflage material characteristics is increasing. Many research results have been achieved based on laboratory hyperspectral instruments. However, most of the existing results only explore the characteristic bands of camouflage materials and have not established a better identification model. Some scholars have proposed a new spectral feature detection method combining mathematical analysis models and statistical calculations with discrimination effect tests and established a physical recognition model in a controlled environment in the laboratory through a large number of experiments. They can get relatively satisfactory results when applied indoors, but they are not fully applicable to data sampled in northern environments.
[0005] Therefore, how to establish an adaptive and invariant recognition model to improve the accuracy of camouflaged target recognition is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention
[0006] In view of this, the present invention provides a method for identifying camouflage materials in northern environments based on spectral characteristic index, establishes an index model with adaptability and invariance, improves the accuracy of camouflage target identification, and is suitable for identifying camouflage nets in northern environments.
[0007] In order to achieve the above object, the present invention provides the following technical solutions:
[0008] A method for identifying camouflage materials in a northern environment based on a spectral characteristic index comprises the following steps:
[0009] Obtain VIS-NIR spectral data of jungle camouflage net samples and VIS-NIR spectral data of typical vegetation in the background environment;
[0010] Resampling the VIS-NIR spectral data of typical vegetation in the background environment;
[0011] The spectral similarity measurement and continuum removal method are used to obtain the spectral reflectance difference between the jungle camouflage net sample and the typical vegetation in the background environment, extract the identification characteristic bands, and construct the ratio camouflage index RCI.
[0012] The recognition threshold condition is set up, and the jungle camouflage net is identified according to the hyperspectral image of the simulated camouflage environment to extract the camouflaged target.
[0013] Optionally, the step of obtaining VIS-NIR spectrum data of the jungle camouflage net sample specifically includes the following steps:
[0014] In a dark room, a spectrometer is used to observe vertically above the jungle camouflage net sample, and a white plate with 100% reflectivity is used as a reference spectrum to obtain the reflectivity data of the jungle camouflage net sample;
[0015] During observation, the jungle camouflage net sample is rotated multiple times and spectral curves are collected multiple times in each direction to obtain a plurality of spectral curves of the jungle camouflage net sample in multiple directions;
[0016] All the acquired spectral curves are subjected to arithmetic averaging to obtain the actual spectral reflectance of the jungle camouflage net sample, which is the VIS-NIR spectral data of the jungle camouflage net sample.
[0017] The technical effect achieved by the above technical solution is: rotating the jungle camouflage net sample multiple times during measurement can reduce the influence of the sample's spectral anisotropy.
[0018] Optionally, after obtaining the VIS-NIR spectrum data of the jungle camouflage net sample, the method further includes:
[0019] The SG smoothing algorithm is used to denoise the VIS-NIR spectral data of the jungle camouflage net sample to obtain preprocessed hyperspectral data, which can improve the quality of the hyperspectral data.
[0020] Optionally, the obtaining of VIS-NIR spectral data of typical vegetation in the background environment is specifically as follows:
[0021] The spectral curves of typical vegetation in the background environment are selected through the ENVI standard spectral library.
[0022] Optionally, the resampling of the VIS-NIR spectral data of the typical vegetation in the background environment is specifically:
[0023] Based on the smoothed VIS-NIR spectral data of the jungle camouflage net sample, that is, the pre-processed hyperspectral data, the VIS-NIR spectral data of the typical vegetation in the background environment are resampled.
[0024] Optionally, constructing the ratio camouflage index RCI specifically includes the following steps:
[0025] The VIS-NIR spectral data of the smoothed jungle camouflage net sample and the resampled VIS-NIR spectral data of the typical vegetation in the background environment were analyzed for similarity measurement. The spectral angle cosine of the jungle camouflage net sample and each typical vegetation was calculated by formula (1), and the similarity comparison between the camouflage net and the zonal vegetation in the common reflectance spectral characteristic band of the green vegetation was obtained.
[0026]
[0027] Where cos(x,y) is the cosine value of the spectral angle, x and y are the reference spectral vector and the target spectral vector respectively;
[0028] The envelope removal analysis was performed on the smoothed VIS-NIR spectrum data of the jungle camouflage net sample and the resampled VIS-NIR spectrum data of the typical vegetation in the background environment. The envelope removal processing was performed on the reflectance spectra of the jungle camouflage net sample and each typical vegetation by formula (2). The difference bands of the camouflage net and the zonal vegetation in the green vegetation were obtained by comparing the processed spectral curves.
[0029]
[0030] In the formula, λ j represents the jth band, R Cj Indicates the envelope removal value of band j, R j represents the original spectral reflectance of band j; R end and R start Respectively represent the original spectral reflectance at the starting point and the end point in the absorption curve; λ end and λ start They represent the starting wavelength and the ending wavelength in the absorption curve respectively; K represents the slope between the starting point band and the ending point band in the absorption curve;
[0031] The spectral characteristics of the jungle camouflage net samples and the typical vegetation in the background environment were analyzed, and the bands with the most obvious absorption characteristic differences were selected. The ratio camouflage index RCI was constructed based on the ratio index for the identification and extraction of camouflaged targets in the northern environment.
[0032] Among them, camouflage nets generally exist in areas with high vegetation coverage. In this environment, the ratio index has better sensitivity. The calculation formula of the ratio index is:
[0033]
[0034] In the formula, R 1 , R 2 is the reflectivity value of the band;
[0035] The calculation formula of the ratio camouflage index RCI is:
[0036]
[0037] In the formula, R 1190 , R 1270 and R 1440 Represent the reflectivity at 1190nm, 1270nm and 1440nm bands respectively.
[0038] Optionally, extracting the disguised target specifically includes the following steps:
[0039] The RCI values of the jungle camouflage net and the typical vegetation in the background environment are calculated respectively, and two threshold conditions that are different from the vegetation are obtained to identify the jungle camouflage net that meets the conditions from the green environment; among them, the identification threshold conditions set are:
[0040]
[0041] Through the hyperspectral imaging experiment, hyperspectral images in a simulated camouflage environment are obtained. A decision tree model is established based on the ratio camouflage index (RCI) and threshold conditions to identify and divide the jungle camouflage nets in the hyperspectral images and extract the camouflaged targets.
[0042] Through the above technical solutions, it can be known that compared with the prior art, the present invention discloses a method for identifying camouflage materials in northern environments based on spectral feature indexes. The spectral differences between camouflage nets and typical zonal vegetation under different humidity conditions are analyzed by spectral data obtained indoors, an index model is constructed, and the identification feature threshold is determined. The imaging hyperspectral data of the simulated environment is used to identify and extract camouflage targets. Compared with the physical model constructed using computer pattern recognition technology, the index model construction method used in the present invention is more adaptable and invariant, and the imaging hyperspectral data is used to verify the recognition model, which reflects a certain practical value and can be applied to the identification of jungle camouflage nets in northern environments. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0044] Figure 1 A flowchart of a method for identifying camouflage materials in northern environments based on spectral characteristic index;
[0045] Figure 2 This is a schematic diagram of a typical northern jungle camouflage net;
[0046] Figure 3 This is a schematic diagram of the spectral reflectance data of the jungle camouflage net sample obtained by measurement and preprocessing;
[0047] Figure 4 It is a schematic diagram for comparing the spectral curves of the reflectance of each vegetation obtained by resampling with the spectral curve of the camouflage net;
[0048] Figure 5 It is a schematic diagram of the spectrum curve after the envelope system is removed;
[0049] Figure 6 Schematic diagram of the pseudo-camouflage environment for the hyperspectral imaging experiment;
[0050] Figure 7 A schematic diagram of building recognition strategy and recognition detection results based on the RCI index. DETAILED DESCRIPTION
[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0052] The embodiment of the present invention discloses a method for identifying camouflage materials in a northern environment based on a spectral characteristic index. Figure 1 As shown, the following steps are included:
[0053] Step 101, obtaining VIS-NIR spectrum data of a jungle camouflage net sample and VIS-NIR spectrum data of typical vegetation in the background environment;
[0054] Step 102, resampling the VIS-NIR spectrum data of typical vegetation in the background environment;
[0055] Step 103, using spectral similarity measurement and continuum removal method to obtain the spectral reflectance difference between the jungle camouflage net sample and the typical vegetation in the background environment, extract the identification characteristic band, and construct the ratio camouflage index RCI;
[0056] Step 104: establish a recognition threshold condition, identify the jungle camouflage net according to the acquired hyperspectral image of the simulated camouflage environment, and extract the camouflaged target.
[0057] Further, in step 101, the obtaining of VIS-NIR spectrum data of the jungle camouflage net sample specifically includes the following steps:
[0058] The VIS-NIR spectrum of the jungle camouflage net sample is obtained by SVC-HR 1024 spectrometer. In this embodiment, the typical jungle camouflage net in the north is used as the experimental object. Figure 2 As shown. The VIS-NIR spectrum measurement was carried out in a dark room, with a 50W standard DC tin wire quartz halogen as the light source, the zenith angle set to 30°, and the distance from the sample surface to 50cm. The reflectance data of the sample was obtained using an American SVC-HR 1024 spectrometer, with a spectral range of 350-2500nm and a field of view of 4°. The sample was observed vertically at 47cm directly above the sample, and a white plate with 100% reflectivity was used as the reference spectrum.
[0059] In addition, in order to reduce the influence of the spectral anisotropy of the sample, the sample was rotated three times during the measurement, each time by about 90°, and the spectral curves of the sample in four directions were obtained. Each direction was collected five times, and a total of 20 sample lines were collected for each sample. The measured spectral reflectance of the camouflage material was obtained after arithmetic averaging. The Savitzky-Golay (SG) filtering algorithm was used to denoise the measured spectral data of the camouflage net sample to obtain the preprocessed hyperspectral data to improve the quality of the hyperspectral data.
[0060] In order to better analyze the spectral characteristics and influencing mechanisms of camouflage materials and simulate the influence of multiple environments on the spectrum, some samples of camouflage materials were immersed in water. The experimental design uses an immersion time of 5 minutes as an interval to measure the spectral curves of the camouflage net under different immersion times. The surface of the camouflage net is saturated with water at around 25 to 35 minutes. At this time, the water attached to the surface of the camouflage net is no longer absorbed, and the influence of the water content on the spectral absorption characteristics reaches the limit. The measured spectral curves of the camouflage net gradually tend to be consistent. Therefore, the spectral curves with immersion time exceeding 30 minutes are eliminated, and the immersion times with significant spectral differences (0min, 5min, 10min, 20min, 30min) are selected to analyze the hyperspectral characteristics of the camouflage net. The actual reflectance spectral curve results of the camouflage net samples obtained by measurement and pretreatment are shown in the following figure. Figure 3 shown.
[0061] Further, in steps 101-102, the VIS-NIR spectral data of typical vegetation in the background environment is obtained, and the VIS-NIR spectral data of typical vegetation in the background environment is resampled, specifically:
[0062] The spectral curves of three vegetation types, namely conifer, deciduous and grass, and the spectral curves of typical vegetation in the coniferous and broad-leaved mixed forests in the northern region, namely poplar (Aspen_Leaf-ADW92-2), pine (Pinon_Pine ANP92-14A needle), and fir (Fir_Tree IH91-2), were selected from the ENVI standard spectral library. The cubic spline interpolation method was used to resample the typical vegetation spectral data based on the smoothed camouflage net spectral data. The comparison between the resampled vegetation reflectance spectral curves and the camouflage net spectral curves is shown in Figure 2. Figure 4 shown.
[0063] Furthermore, in step 103, the construction of the ratio camouflage index RCI specifically includes the following steps:
[0064] The similarity measurement analysis is performed on the VIS-NIR spectral data of the smoothed jungle camouflage net sample and the VIS-NIR spectral data of the typical vegetation in the background environment after resampling:
[0065] "Green reflection peak", "red edge", "near infrared plateau" and "water absorption band" are common features of green vegetation reflectance spectra, which provide basic spectral parameters for detecting spectral anomalies in green vegetation environments. The spectral angle cosine of the jungle camouflage net sample and each typical vegetation in these four characteristic parameter bands is calculated by formula (1), and the similarity comparison between the camouflage net and the zonal vegetation in the common reflectance spectral characteristic bands of green vegetation is obtained;
[0066]
[0067] Where cos(x,y) is the cosine value of the spectral angle, x and y are the reference spectral vector and the target spectral vector respectively;
[0068] The envelope removal analysis was performed on the smoothed VIS-NIR spectral data of the jungle camouflage net sample and the resampled VIS-NIR spectral data of the typical vegetation in the background environment:
[0069] The envelope of the reflectance spectrum of the jungle camouflage net sample and each typical vegetation is removed by equation (2). The difference bands of the camouflage net and the zonal vegetation in the green vegetation are obtained by analyzing and comparing the spectral curves after the processing. The spectral curve after the envelope is removed is as follows: Figure 5 As shown;
[0070]
[0071] In the formula, λ j represents the jth band, R Cj Indicates the envelope removal value of band j, R j represents the original spectral reflectance of band j; R end and R start Respectively represent the original spectral reflectance at the starting point and the end point in the absorption curve; λ end and λ start They represent the starting wavelength and the ending wavelength in the absorption curve respectively; K represents the slope between the starting point band and the ending point band in the absorption curve;
[0072] The difference bands between the camouflage net and various vegetation extracted in the study are sensitive bands for camouflage net identification. Compared with vegetation, the camouflage net (dry) has no obvious absorption characteristics at 970nm and 1190nm, and has a wide absorption valley at 1440nm, and the two have a large degree of distinction. In the 900-1300nm band, the camouflage net's spectral curve drops rapidly; in the 1300-1600nm band, the camouflage net's spectral curve is at a wide absorption valley, fluctuates steadily, and has a small slope; while the spectral curve of vegetation fluctuates greatly, and there are steep rises and falls in the 1150-1300nm and 1300-1440nm bands, respectively, with a large slope. Through analysis, in the near-infrared band, near 970nm, 1190nm and 1440nm, the camouflage net (dry) and vegetation reflectance spectra are significantly different.
[0073] Analyze the difference in spectral reflectance between the camouflage net and typical vegetation in the background environment, and construct the ratio camouflage index RCI:
[0074] Camouflage nets are generally found in areas with high vegetation coverage. In this environment, the ratio index has better sensitivity. Formula (3) is the calculation formula for the ratio index:
[0075]
[0076] In the formula, R 1 , R 2 is the reflectivity value of the band;
[0077] The spectral characteristics of the jungle camouflage net sample and the typical vegetation in the background environment were analyzed, and the 1150-1440nm band with obvious absorption characteristics was selected. The ratio camouflage index RCI was constructed based on the ratio index for the recognition and extraction of camouflaged targets in the northern environment, as shown in formula (4):
[0078]
[0079] In the formula, R 1190 , R 1270 and R 1440 Respectively represent the reflectivity at 1190nm, 1270nm and 1440nm bands. Among them, the selection of 1190nm and 1270nm bands corresponds to the characteristic bands at the trough and right shoulder peak of the vegetation absorption valley that is different from the camouflage net in the range of 1150-1300nm; the selection of 1270nm and 1440nm bands corresponds to the characteristic bands at the left shoulder peak and trough of the vegetation absorption valley that is different from the camouflage net in the range of 1300-1440nm.
[0080] Furthermore, in step 104, extracting the disguised target specifically includes the following steps:
[0081] The ratio camouflage index model established through the above steps is used to calculate the RCI values of the jungle camouflage net and the typical vegetation in the background environment. 1 The values of are all greater than 1.4, that is, the reflectivity slope of the camouflage net is greater than 1.4; in the 1300-1440nm band RCI 2 Both are less than 1.4, that is, the reflectivity slope of the camouflage net is less than 1.4; both are different from vegetation. If these two conditions can be met at the same time, the camouflage net can be identified from the green environment; therefore, the identification threshold condition is set as:
[0082]
[0083] In order to further verify the practicality of the index, it is applied to hyperspectral images. Using a simulation model containing vegetation, gravel, rocks, vehicles and other objects, the vehicle is camouflaged with a jungle camouflage net with vegetation as the background target, and the Image-λ "spectral image" series hyperspectral camera is used to conduct hyperspectral imaging experiments on the simulated environment before and after camouflage. Figure 6 As shown (the top is a hyperspectral imager, the lower left is a non-camouflaged environment, and the lower right is a simulated camouflaged environment).
[0084] A decision tree model was established based on the ratio camouflage index (RCI) and threshold conditions to identify and divide the jungle camouflage net in the hyperspectral image. The results are as follows: Figure 7 As shown in the figure. In the recognition feature extraction results, the white part represents the camouflage net, and the black part represents other background objects. From the results, it can be seen that the extracted camouflage net is basically consistent with the original image in shape and size. The recognition detection accuracy of the experimental results calculated by intersection ratio can reach 0.95, indicating that the recognition effect is relatively good and the classification accuracy is high.
[0085] In view of the problems existing in the prior art, the present invention analyzes the spectral differences between camouflage nets and typical zonal vegetation under different humidity conditions through spectral data obtained indoors, constructs an exponential model, determines the recognition feature threshold, and performs recognition and extraction of camouflage targets on the imaging hyperspectral data of the simulated environment. Compared with the physical model constructed using computer pattern recognition technology, the exponential model construction method used in the present invention is more adaptable and invariant, and the recognition model is verified using imaging hyperspectral data, which reflects a certain practical value and can be applied to the recognition of jungle camouflage nets in northern environments.
[0086] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for identifying camouflage materials in northern environments based on spectral characteristic index, characterized in that: The following steps are involved: Obtain VIS-NIR spectral data of the jungle camouflage net sample and VIS-NIR spectral data of typical vegetation in the background environment, specifically, obtain the actual spectral reflectance of the jungle camouflage net sample, wherein the spectral range of the reflectance data is 350-2500nm; Resampling the VIS-NIR spectral data of typical vegetation in the background environment; The spectral similarity measurement and continuum removal method are used to obtain the spectral reflectance difference between the jungle camouflage net sample and the typical vegetation in the background environment, extract the identification characteristic bands, and construct the ratio camouflage index RCI. Establish recognition threshold conditions, identify the jungle camouflage net based on the acquired hyperspectral image of the simulated camouflage environment, and extract the camouflaged target; The method of obtaining the ratio camouflage index RCI includes: Through spectral similarity measurement and envelope removal processing, the spectral characteristics of the jungle camouflage net sample and the typical vegetation in the background environment were analyzed, and the bands with the most obvious absorption characteristic differences were selected to construct the ratio camouflage index RCI based on the ratio index. The calculation formula of the ratio index is: In the formula, R1 and R2 are the reflectivity values of the band; The calculation formula of the ratio camouflage index RCI is: In the formula, R 1190 , R 1270 and R 1440 Respectively represent the reflectivity at 1190nm, 1270nm and 1440nm bands; Methods for extracting camouflaged targets include: Calculate the RCI values of the jungle camouflage net and the typical vegetation in the background environment respectively, and obtain two threshold conditions that are different from the vegetation to identify the jungle camouflage net that meets the conditions from the green environment; Through the hyperspectral imaging experiment, hyperspectral images in a simulated camouflage environment are obtained. A decision tree model is established based on the ratio camouflage index (RCI) and threshold conditions to identify and divide the jungle camouflage nets in the hyperspectral images and extract the camouflaged targets.
2. The method for identifying camouflage materials in northern environments based on spectral characteristic index according to claim 1, characterized in that: The method of obtaining the VIS-NIR spectrum data of the jungle camouflage net sample specifically comprises the following steps: In a dark room, a spectrometer is used to observe vertically above the jungle camouflage net sample, and a white plate with 100% reflectivity is used as a reference spectrum to obtain the reflectivity data of the jungle camouflage net sample; During observation, the jungle camouflage net sample is rotated multiple times and spectral curves are collected multiple times in each direction to obtain a plurality of spectral curves of the jungle camouflage net sample in multiple directions; All the acquired spectral curves are subjected to arithmetic averaging to obtain the actual spectral reflectance of the jungle camouflage net sample, which is the VIS-NIR spectral data of the jungle camouflage net sample.
3. The method for identifying camouflage materials in northern environments based on spectral characteristic index according to claim 1, characterized in that: After obtaining the VIS-NIR spectrum data of the jungle camouflage net sample, the method further includes: The SG smoothing algorithm is used to perform denoising on the VIS-NIR spectral data of the jungle camouflage net sample to obtain preprocessed hyperspectral data.
4. The method for identifying camouflage materials in northern environments based on spectral characteristic index according to claim 1, characterized in that: Obtain VIS-NIR spectral data of typical vegetation in the background environment, specifically: The spectral curves of typical vegetation in the background environment are selected through the ENVI standard spectral library.
5. The method for identifying camouflage materials in northern environments based on spectral characteristic index according to claim 3 is characterized in that: The resampling of the VIS-NIR spectral data of the typical vegetation in the background environment is specifically as follows: Based on the smoothed VIS-NIR spectral data of the jungle camouflage net sample, that is, the pre-processed hyperspectral data, the VIS-NIR spectral data of the typical vegetation in the background environment are resampled.
6. The method for identifying camouflage materials in northern environments based on spectral characteristic index according to claim 3, characterized in that: In the step of constructing the ratio camouflage index RCI, the similarity measurement is: performing similarity measurement analysis on the VIS-NIR spectrum data of the smoothed jungle camouflage net sample and the VIS-NIR spectrum data of the typical vegetation in the background environment after resampling, calculating the spectral angle cosine of the jungle camouflage net sample and each typical vegetation by formula (3), and obtaining the similarity comparison between the camouflage net and the zonal vegetation in the common reflectance spectral characteristic band of the green vegetation; Where cos(x,y) is the cosine value of the spectral angle, x and y are the reference spectral vector and the target spectral vector respectively; The envelope removal analysis is as follows: the VIS-NIR spectrum data of the smoothed jungle camouflage net sample and the VIS-NIR spectrum data of the typical vegetation in the background environment after resampling are subjected to envelope removal analysis. The reflectance spectra of the jungle camouflage net sample and each typical vegetation are subjected to envelope removal processing by formula (4). The difference bands of the camouflage net and the zonal vegetation in the green vegetation are obtained by comparing the processed spectral curves. In the formula, λ j represents the jth band, R Cj Indicates the envelope removal value of band j, R j represents the original spectral reflectance of band j; R end and R start Respectively represent the original spectral reflectance at the starting point and the end point in the absorption curve; λ end and λ start They represent the starting point wavelength and the ending point wavelength in the absorption curve respectively; K represents the slope between the starting point band and the ending point band in the absorption curve.
7. The method for identifying camouflage materials in northern environments based on spectral characteristic index according to claim 6, characterized in that: The identification threshold condition for identifying the jungle camouflage net that meets the conditions from the green environment is: