CN103383348B - Vegetation-covered area high-spectrum remote-sensing altered mineral extracting method - Google Patents
Vegetation-covered area high-spectrum remote-sensing altered mineral extracting method Download PDFInfo
- Publication number
- CN103383348B CN103383348B CN201310204230.5A CN201310204230A CN103383348B CN 103383348 B CN103383348 B CN 103383348B CN 201310204230 A CN201310204230 A CN 201310204230A CN 103383348 B CN103383348 B CN 103383348B
- Authority
- CN
- China
- Prior art keywords
- spectrum
- vegetation
- plant
- wave
- abnormality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001228 spectrum Methods 0.000 title claims abstract description 83
- 229910052500 inorganic mineral Inorganic materials 0.000 title claims abstract description 42
- 239000011707 mineral Substances 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 26
- 239000002689 soil Substances 0.000 claims abstract description 27
- 230000003595 spectral effect Effects 0.000 claims abstract description 24
- 230000002159 abnormal effect Effects 0.000 claims abstract description 23
- 230000005856 abnormality Effects 0.000 claims abstract description 18
- 238000010586 diagram Methods 0.000 claims abstract description 15
- 238000012937 correction Methods 0.000 claims abstract description 14
- 238000000513 principal component analysis Methods 0.000 claims abstract description 11
- 230000005855 radiation Effects 0.000 claims abstract description 10
- 238000000985 reflectance spectrum Methods 0.000 claims abstract description 8
- 238000006243 chemical reaction Methods 0.000 claims abstract description 6
- 238000012360 testing method Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000002310 reflectometry Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000005764 inhibitory process Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000010183 spectrum analysis Methods 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 5
- 239000003086 colorant Substances 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000008439 repair process Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 16
- 230000004075 alteration Effects 0.000 abstract description 15
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000000926 separation method Methods 0.000 abstract description 4
- 238000005457 optimization Methods 0.000 abstract 1
- 235000010755 mineral Nutrition 0.000 description 37
- 241000196324 Embryophyta Species 0.000 description 24
- 239000011435 rock Substances 0.000 description 14
- 230000009466 transformation Effects 0.000 description 8
- 230000000694 effects Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 230000033558 biomineral tissue development Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 239000007943 implant Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 244000274847 Betula papyrifera Species 0.000 description 1
- 235000009113 Betula papyrifera Nutrition 0.000 description 1
- 235000009109 Betula pendula Nutrition 0.000 description 1
- 235000010928 Betula populifolia Nutrition 0.000 description 1
- 235000002992 Betula pubescens Nutrition 0.000 description 1
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 240000000604 Lespedeza bicolor Species 0.000 description 1
- 235000016677 Lespedeza bicolor Nutrition 0.000 description 1
- 244000179560 Prunella vulgaris Species 0.000 description 1
- 241001480055 Quercus mongolica Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 235000005272 common selfheal Nutrition 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000012847 principal component analysis method Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 239000011573 trace mineral Substances 0.000 description 1
- 235000013619 trace mineral Nutrition 0.000 description 1
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention relates to a kind of vegetation-covered area high-spectrum remote-sensing altered mineral extracting method.The page aiming at wide leaf plant with detecting gun directly measures blade face reflectance spectrum, and herbaceous plant measures the reflectance spectrum of canopy, its reflectance spectrum of fresh of soil testing; Hyperion data are not calibrated and are subject to the removal of steam image wave band, the conversion of absolute radiation value, bad line reparation and atmospheric correction; According to principal component analysis (PCA) eigenwert, find out comparatively large and two principal components that symbol is contrary of absolute value, do 2-D scatter diagram according to two principal components; Again according to scatter diagram delineation abnormality.By the comparison to above-mentioned spectral characteristic matching method and mineral extraction model, adopt the multi-layer information separation of background, interference, abnormal information, focus on selecting best wave band to carry out space characteristics analysis.By analyzing high-spectral data feature, vegetation-covered area regional feature, Optimal Bands Selection used during optimization space characteristics, better suppresses vegetation information, strengthens the feature of alteration Information.
Description
Technical Field
The invention relates to application of a remote sensing image in geology, in particular to a method for extracting altered minerals by utilizing a hyperspectral remote sensing image in a vegetation coverage area.
Background
The hyperspectral remote sensing has very continuous wave bands, the spectral resolution reaches the nanometer level, different types of minerals can be identified through micro spectral differences among different minerals, and the existing hyperspectral remote sensing mineral qualitative identification is well verified in rock bare areas through a spectral hourglass model based on spectral similarity and wave band decorrelation. However, in a vegetation coverage area, rocks and minerals are represented as weak information and small targets on a remote sensing image and are difficult to identify, and how to extract the weak information by using remote sensing data is a difficult problem of the modern remote sensing geological application technology.
The method has the defects that the method has no universality, needs to be respectively statistically analyzed in different areas, and has large calculation error by utilizing the earth surface parameters because the remote sensing data are all mixed pixels. In addition, a principal component analysis method, a spectral de-envelope matching method, and the like, which are commonly used for remote sensing data processing, are greatly affected by vegetation.
Cross A P, Review in Economic geography, 2009,16:59-82. Cross and Moore develops a Feature-Oriented Principal component selection (Feature Oriented Principal component selection) technology aiming at the characteristics of vegetation, soil development and limited geological map in tropical rainforest zones of Brazil and the defects that a Standard Principal component method (Standard PCA) cannot correspond a mineral spectrum with Principal component transformation. The method is successfully applied to the alteration information mapping in regions such as the mineralization alteration rock mapping in a brazilian tropical rainforest region, the porphyry copper mineralization alteration rock mapping in an Iran vegetation region and the like.
"mineral science report 2011, 922 one 926, of Chanting et al," application of hyperspectral remote sensing in porphyry ore deposit alteration information extraction ", in extracting porphyry ore deposit alteration information by using hyperspectrum, extracting alteration information of a research area by using measured spectral data of the research area through a spectral feature absorption position matching method, and classifying and sorting the alteration information by combining a geological mineralization zone of the research area to complete a hyperspectral model of deep alteration of the ore area.
"national resource science and technology management" 2012,29(6),84-87 "change mineral information and construct in the application of the hyperspectral remote sensing prospecting in the beginning" in the mineral extraction model MNF transform, PPI index calculate pure end member "in the HyMap remote sensing image through the Spectrum Angle (SAM) classification success extraction mineral, prove this method is suitable for the universality fully. The method has good extraction effect in bare rock areas, but in a vegetation coverage area, the quantity of 'pure pixels' calculated during PPI pure pixel calculation is very small, and the PPI pure pixels contain vegetation information.
Ma Chaofei et al geology sciences 2002,37(3): 365-: the method comprises the steps of establishing a relation among extracted components, spectrums and characteristic wave bands by researching corresponding relations of first-order derivatives of different plant trace elements and plant spectrums and combining rock and soil geochemical data, plant leaf imaging data, biochemical component data and the like, and dividing large-area abnormal information of a covered area of a plant in a remote sensing image to obtain a good map filling effect.
RowinFi et al ' spectroscopy and spectral analysis ' 2010 (30) ' 1628-. This approach is aimed at being suitable when the mineral component is low.
Zhang Yuanfei et al, national soil resource remote sensing 2011,04, 6-13, "multilayer separation technology model and application research of remote sensing alteration information", when multispectral remote sensing alteration information extraction is researched, a multilayer separation technology of a background and an interference is intensively researched, and various rules about the background, the interference and alteration abnormity are summarized. The method of the article is characterized in that the optimal wave band in remote sensing data is selected firstly, the data is projected in a two-dimensional space, and through the spatial distribution of different substances in a spectral feature space, interference is removed as far as possible to leave alteration abnormal information. The interference-free mineral extraction method uses multispectral and hyperspectral data, and has good effects in low vegetation coverage areas and bare rock areas. When the hyperspectral data is used, the interval of the wave bands of the hyperspectral data is dense, the characteristic difference of the data in a two-dimensional space is small, and the extraction effect is not as good as that of the multispectral data. In the vegetation coverage area, the extracted alteration abnormality contains vegetation information which cannot be effectively removed. The hyperspectral data is used in vegetation areas, and the extraction effect is relatively poor.
In conclusion, the existing method model mostly adopts experience and one-by-one experiment to select wave bands, has strong randomness, lacks the basis of wave band selection reference, and cannot well separate the weak information of minerals in the high-reflection vegetation background without thoroughly inhibiting the vegetation spectrum.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a vegetation coverage area hyperspectral remote sensing altered mineral extraction method based on vegetation inhibition.
The purpose of the invention is realized by the following technical scheme:
1. a method for extracting altered minerals in a plant covered area through hyperspectral remote sensing is characterized by comprising the following steps:
a. in a time period that the solar altitude angle is larger than 45 degrees, firstly measuring the reflectivity of the white board, and observing a spectrum curve of the white board until the curve is approximate to a straight line near 100% reflectivity, and finishing correction;
b. then, a detection gun is aligned to the page of the wide-leaf plant to directly measure the reflection spectrum of the leaf surface, the herbaceous plant measures the reflection spectrum of the canopy, the soil tests the reflection spectrum of the fresh surface of the soil, and the spectrum curve of the tested plant and soil is stored;
c. comparing the measured plant reflectance spectrum with the soil reflectance spectrum, selecting a pair of different wave bands with larger difference between the plant reflectance and the soil reflectance, wherein the plant reflectance of the two wave bands is the same, and each two selected wave bands are used as a pair;
d. preprocessing EO-1Hyperion data, including uncalibrated and water vapor image wave band removal, absolute radiation value conversion, bad line repair and atmospheric correction of the Hyperion data;
e. finding out the corresponding wave band in step c from the preprocessed data, and for each wave bandAnd (3) performing band difference calculation on all bands: calculating the formula: b = b1-b2
In the formula: b is new spectrum emissivity calculated by wave band difference, b1、b2(ii) for each selected pair, implanting the suppressed spectral reflectance;
f. the wave bands after vegetation inhibition are taken as a group, and the formula is as follows:
eigenvalue and eigenvector of covariance matrix
|λE-C|=0 ②
A=CT ③
Y=AX ④
Wherein x isi(k, l) is the value of the row number (k, iota) at the i-band where the pixel point is located,is the data average value of the i wave band, C (i, j) is a covariance matrix, A is a transposed matrix (T) of the covariance matrix C of the X space, Y is a new principal component matrix, and X is an original matrix.
And performing principal component analysis to obtain principal components. After the principal components are found, according to the formula:
obtaining a principal component analysis characteristic value;
wherein alpha isiIs the characteristic value of the nth principal component of the original variable X, λiRepresents the nth eigenvalue of the covariance matrix C, these two n being identical;
g. according to the principal component analysis characteristic values, two principal components with larger absolute values and opposite signs are found out, and a 2-D scatter diagram is made according to the two principal components;
h. selecting data at the inflection points of a scatter diagram as a defined end-member spectrum, marking each inflection point as different colors, when one inflection point needs to determine a range, firstly defining the outermost layer of anomalies as primary anomalies, generally expressed by red, and according to a formula:
wherein,the average spectrum value is obtained, q … … N is the number of pixels, and Xq is the spectrum curve of the q-th pixel to obtain the average spectrum of the first-order abnormality.
Secondly, the second-level abnormality is circled inwards, usually expressed by green, and the average spectrum of the second-level abnormality is obtained through a formula; and then performing spectrum matching on the two spectrums, judging the spectrums to be of the same type if the score is more than 0.8, and indicating that the primary abnormality and the secondary abnormality at the inflection point are of the same type, otherwise, judging the abnormalities to be of different types.
i. After all abnormal information is selected, the spectrum matching is carried out through the USGS standard spectrum library, and the mineral type of the abnormal information can be determined through the processing of a spectrum analysis tool in ENVI software during the spectrum matching.
Has the advantages that: due to the influence of vegetation, exposed rock soil is mostly a small-area in a vegetation coverage area, the proportion of the exposed rock soil in the remote sensing image is small, and the exposed rock soil is expressed as small target and weak information. Therefore, in vegetation coverage areas, particularly high vegetation coverage areas, due to the high reflectivity effect of the coverage vegetation of rock soil, the method for extracting the altered minerals in the bare rock areas directly carried over cannot directly extract the altered minerals. Even if the extracted mineral information is mostly distributed along roads, rivers and other relatively low coverage areas, the effect is relatively poor in bare areas due to the influence of high-reflectivity vegetation of adjacent pixels. Aiming at the problem of extraction of the altered mineral in the vegetation coverage area by hyperspectral remote sensing, the invention is characterized in that a method for selecting the optimal hyperspectral band combination to carry out vegetation spectrum inhibition is provided based on the actual measurement spectrum analysis of regional rocks, soil and vegetation, the technical problem of separation of vegetation as background and interference and abnormal information is solved by the spatial structure characteristic analysis of the selected band, and the extraction of the altered mineral information in the vegetation coverage area is realized by combining spectrum matching. Compared with the existing mineral extraction model, the rock and mineral types of the area are obtained. According to the invention, through the ground actual measurement spectral feature analysis of the vegetation coverage area, the optimal waveband combination is selected, the vegetation information is better inhibited, the alteration mineral information feature under the vegetation coverage is effectively enhanced and extracted, and a corroboration is provided for the search and prediction of mineral resources.
Description of the drawings:
FIG. 1 is a flow chart of a method for extracting altered minerals through hyperspectral remote sensing of vegetation coverage areas.
FIG. 2 is a graph of measured plant and measured soil spectra.
FIG. 3 is a comparison of the measured plant spectrum and soil spectrum.
FIG. 4 is a comparison graph of spectral curves of vegetation before and after atmospheric correction
(a) Before atmospheric correction; (b) after atmospheric correction.
FIG. 5 is a diagram showing an analysis of abnormality in a scattergram.
FIG. 6 is a scatter plot showing all anomalies and the distribution of anomalies
(a) A spectrum curve corresponding to the abnormal information; (b) all anomaly information of the scatter diagram;
(c) and the position of the abnormal information in the remote sensing image.
Fig. 7 is a graph of mineral extraction results.
The specific implementation mode is as follows:
the following detailed description is made with reference to the accompanying drawings and examples:
the working flow of the method for extracting the hyperspectral remote sensing altered mineral of the vegetation coverage area is shown in figure 1, and in order to better explain the hyperspectral remote sensing mineral identification method based on the improved vegetation inhibition method, the following mineral identification is carried out by taking hyperspectral data of the Huma area of Heilongjiang river as an example, and the specific steps are as follows:
collecting plant and soil samples in the Huma area of Heilongjiang for spectrum test.
a. And (3) correcting the instrument: the reflectivity of the white board is measured by a FieldSpec Pro portable spectral radiation spectrometer (the wave band is 350-2500 nm) produced by American ASD company in a time period when the solar altitude is more than 45 degrees, and the spectrum curve is observed until the curve is approximately a straight line near 100 percent reflectivity.
b. Measurement of a spectral curve: after correction, a detection gun is used for aiming at the plant, the reflection spectrum curve of the target plant is measured, and finally, the spectrum curve is stored. Directly measuring the leaf surface reflectance spectrum of broad-leaf plants such as quercus mongolica, white birch, lespedeza bicolor and the like; and the common selfheal fruit-spike is a herbaceous plant, and the reflection spectrum of the canopy is measured; the soil was planed to a fresh surface and the spectrum was tested as shown in figure 2.
Then, the measured spectra of all plants (10 points in total, 50 spectra) were averaged, and the measured spectra of all soils (10 points in total, 12 spectra) were averaged as the spectrum to be used in the next step.
c. Selecting a wave band: comparing the measured plant spectrum with the soil spectrum (fig. 3), selecting two bands with larger difference between the plant reflectivity and the soil reflectivity, and simultaneously, the plant reflectivity values of the two bands are the same; every two bands thus selected as a pair, such as: the spectral reflectance of the plant with the wavelength of 2314nm is the same as the reflectance of the plant with the wavelength of 701nm, but the difference of the soil reflectance of the two wavelengths is large. In this experiment, six pairs of bands were identified (see Table 1).
TABLE 1 selection of 6 pairs of bands for implant suppression
d. EO-1Hyperion data preprocessing: the EO-1Hyperion data are preprocessed, and the preprocessing process comprises the steps of carrying out uncalibrated and water vapor image wave band removal, absolute radiation value conversion, bad line repair, atmospheric correction and the like on the Hyperion data. The specific pretreatment is as follows:
firstly, removing the uncalibrated and water-gas image wave band
In the 242 wave bands of the Hyperion image, the wave bands of radiometric calibration are VNIR 8-57 and SWIR 77-224; the uncalibrated wave bands are set to be 0 values, and are wave bands 1-7, wave bands 58-76 and wave bands 225-242. The scaled bands are extracted to produce a 198 band image.
As the VNIR bands 56-57 and SWIR bands 77-78 overlap, there are actually only 196 separate bands. The noise of the SWIR 77-78 is larger than that of the VNIR 56-57, and the VNIR 56-57 is usually retained, while the SWIR 77-78 is deleted. Thus, a 196 band image is generated, which corresponds to the original bands 8-57 and 79-224.
In the hyperspectral remote sensing data, the spectral ranges of 1356-1417 nm, 1820-1932 nm and the wave bands larger than 2395nm are greatly influenced by water vapor, and in the wave bands, the ground information is rarely contained. Therefore, they need to be eliminated. For Hyperion data, the wave bands influenced by water vapor are 121-127, 167-178 and 224, and the total wave bands are 20. 176 wave bands, namely 8-57, 79-120, 128-166 and 179-223 (the number represents the wave band) are left after the wave bands influenced by the steam are removed. The wavelength ranges of the rejected bands are shown in Table 2.
TABLE 2 rejected Hyperion bands
| Hyperion original wave band | Wavelength range (nm) |
| 1～7 | 355～416 |
| 58～78 | 936～923 |
| 121～127 | 1356～1426 |
| 167～178 | 1820～1931 |
| 224～240 | 2395～2577 |
Conversion of pixel value to absolute radiation value
The L1 product data set of Hyperion is recorded by signed integer data, and the numerical range is-32767- + 32767. However, in practice, the radiation value of the ground object is very small, the maximum radiation of VNIR is 750W/m2/sr/Lm, and the maximum radiation of SWIR is 350W/m 2/sr/Lm. Therefore, when the product is produced, the scale factors are used, and the factor coefficients of VNIR and SWIR are 40 and 80, respectively.
In analyzing application Hyperion data, the pixel values must be converted to absolute radiance values. First, all VNIR bands are divided by 40 to produce a new image file, and all SWIR bands are divided by 80 to produce another new image file. Then, the two image files are combined to obtain an absolute radiation value image.
③ repairing damaged wire
Since there is some error in the calibration of the sensors, there is still abnormal data in the L1 grade product of Hyperion. A row or column with no data or very small data values is often referred to as a bad line. The Hyperion images of 176 wave bands are checked wave band by band, and the wave bands with bad lines and the corresponding column numbers are recorded. Then, the average of its adjacent rows or columns is used for repair.
Data format conversion and atmospheric correction
The purpose of atmospheric correction is to eliminate the influence of factors such as atmosphere and illumination on the reflection of the ground objects and obtain the real physical model parameters of the ground surface, such as the reflectivity, radiance, ground surface temperature and the like of the ground objects. The extraction of the altered mineral information by using the Hyperion image needs to acquire the real reflectivity of the earth surface, so that the data needs to be corrected by atmosphere, and the atmospheric correction is performed by adopting an ENVI FLAASH module. FIG. 4 is a comparison graph of the spectrum curves of vegetation before and after atmospheric correction.
e. Calculating the difference of the remote sensing image wave bands: finding out the corresponding wave band in the table 1 in the data after EO-1Hyperion pretreatment, and performing differential calculation by using the wave band obtained in the step c, wherein the formula of the differential calculation is as follows: b = B1-B2, wherein: b is the new spectral reflectance image calculated from the band difference, B1, B2 are selected for each pair to be used to implant the suppressed spectral reflectance image. In the examples as follows: the first set of difference calculations B1= B2314nm-B701nm, i.e. the difference between the reflectance image with a center wavelength of 2314nm and the reflectance image with a center wavelength of 701 nm. B2= B2203nm-B681nm, i.e., the difference between the reflectance image with the center wavelength of 2203nm and the reflectance image with the center wavelength of 681 nm. The number of the new vegetation inhibited bands in this experiment was 6.
The substitution formula is calculated as:
B1=2314-701
B2=2203-681
B3=2183-671
B4=2072-548
B5=2213-691
B6=1336-742
f. and (3) main component analysis: the 6 new bands (6 bands obtained in the above equation) obtained in step e are grouped, and principal component analysis is performed according to the following equation to obtain 6 new principal components and principal component feature values (see table 3). That is, the value of the B1 image obtained in (e) is one column, the value of the B2 image is the second column, and so on, forming a matrix of 6 columns and rows of image values, denoted as X, and substituting X into the matrix,
The principal component analysis transformation process is implemented using the principal component transformation function in ENVI.
TABLE 3 principal component analysis characteristic value Table
g. 2-D scattergram: from the principal component analysis eigenvalues (table 3), two principal components having larger absolute values and opposite signs were found as 2-D scattergrams, in which the value of PC1 was negative and the value of PC2 was positive, and the absolute values of PC1 and PC2 were larger, so PC1 and PC2 were used as scattergrams (fig. 5).
h. Abnormal range delineation: firstly, in a 2-D scatter diagram, data at inflection points of the scatter diagram are selected as defined end-member wave spectrums, each inflection point is marked with different colors, and one inflection point is abnormal information of one type.
When the range of the abnormal information is determined, firstly, the outermost layer of abnormal information is determined to be a first-level abnormal (red), the average spectrum (red) of the abnormal information is obtained, and the formula is as follows:
is the average spectral value, q … … N is the number of pixels, Xq is the spectral curve of the qth pixel; the second-order anomaly (green) was further determined and the average spectrum (green) was obtained.
Carrying out spectrum matching on the red and green spectrums, and processing the spectrums through ENVI software when the spectrums are matched, wherein the steps comprise spectrum characteristic fitting processing and final scoring; if the score is more than 0.8, the spectrum is judged to be of the same type, and the first-order abnormality and the second-order abnormality at the inflection point are shown to be of the same type (as shown in FIG. 5).
And fourthly, sequentially delineating all inflection points (abnormal information) in the 2-D scatter diagram, marking different inflection points as different colors, and obtaining the average spectrum of each inflection point. By defining the range of values of PC2 and PC1 of the scatter diagram, the distribution range of the abnormal information can be displayed in the remote sensing image at the same time. The average spectrum at each inflection point, i.e. the average spectrum of the pixels displaying abnormal information distribution in the remote sensing image, is respectively obtained (fig. 6). And after the inflection point is circled on the scatter diagram, marking out the corresponding range of the inflection point in the scatter diagram on the preprocessed remote sensing data image. Each pixel on the remote sensing image has a spectrum curve, and the spectra of all pixels in the range are averaged. The average curve is the average spectrum curve of the inflection point on the scatter diagram.
i. Matching a standard spectrum library: and h, determining the average spectrum by using the spectrum abnormal information determined in the step h, wherein the mineral spectrum (belonging to the USGS spectrum library) in the standard spectrum library as a reference spectrum, judging by using spectral analysis (including a spectrum angle SAM, a spectrum characteristic fitting SFF and a binary code, wherein the weights are respectively 0.3, 0.4 and 0.3) in ENVI software, and finally determining the mineral type of the abnormal information when the average spectrum curve is similar to the mineral curve in the USGS standard spectrum library, the higher the score is, the highest matching score is the abnormal mineral type, so as to obtain a mineral extraction result (as shown in figure 7).
Claims (1)
1. A method for extracting altered minerals in a plant covered area through hyperspectral remote sensing is characterized by comprising the following steps:
a. in a time period that the solar altitude angle is larger than 45 degrees, firstly measuring the reflectivity of the white board, and observing a spectrum curve of the white board until the curve is approximate to a straight line near 100% reflectivity, and finishing correction;
b. then, a detection gun is aligned to the leaf surface of the wide-leaf plant to directly measure the leaf surface reflection spectrum, the herbaceous plant measures the reflection spectrum of the canopy, the soil tests the reflection spectrum of the fresh surface of the soil, and the wave spectrum curve of the tested plant and soil is stored;
c. comparing the measured plant reflectance spectrum with the soil reflectance spectrum, selecting a pair of different wave bands with larger difference between the plant reflectance and the soil reflectance, wherein the plant reflectance of the two wave bands is the same, and each two selected wave bands are used as a pair;
d. preprocessing EO-1Hyperion data, including uncalibrated and water vapor affected wave band removal, absolute radiation value conversion, bad line repair and atmospheric correction of the Hyperion data;
e. finding out the corresponding wave bands in the step c from the preprocessed data, and carrying out wave band difference calculation on each pair of wave bands: calculating the formula: b = b1-b2
In the formula: b is the new spectral reflectivity calculated from the difference of the wave bands, b1、b2(ii) for each selected pair, implanting the suppressed spectral reflectance;
f. taking the wave bands after vegetation inhibition as a group, and solving a principal component analysis characteristic value through a covariance matrix formula;
g. according to the principal component analysis characteristic values, two principal components with larger absolute values and opposite signs are found out, and a 2-D scatter diagram is made according to the two principal components;
h. selecting data at the inflection points of a scatter diagram as a defined end-member spectrum, marking each inflection point as different colors, when one inflection point needs to determine a range, firstly defining the outermost layer of anomalies as primary anomalies, generally expressed by red, and according to a formula:
⑥
calculating the average spectrum of the first-order abnormity;
secondly, the second-level abnormality is circled inwards, usually expressed by green, and the average spectrum of the second-level abnormality is obtained through a formula; then, matching the two wave spectrums, judging the wave spectrums to be of the same type if the fraction is more than 0.8, and showing that the primary abnormality and the secondary abnormality at the corner are of the same type, otherwise, the first abnormality and the second abnormality are of different typesAn abnormality in whichIs the average spectral value, q … … N is the number of pixels,is the spectral curve of the qth pixel;
i. and after all abnormal information is selected, performing spectrum matching through a USGS standard spectrum library, and determining the mineral type of the abnormal information through processing by a spectrum analysis tool in ENVI software during the spectrum matching.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310204230.5A CN103383348B (en) | 2013-05-28 | 2013-05-28 | Vegetation-covered area high-spectrum remote-sensing altered mineral extracting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310204230.5A CN103383348B (en) | 2013-05-28 | 2013-05-28 | Vegetation-covered area high-spectrum remote-sensing altered mineral extracting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103383348A CN103383348A (en) | 2013-11-06 |
| CN103383348B true CN103383348B (en) | 2015-09-30 |
Family
ID=49491197
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310204230.5A Active CN103383348B (en) | 2013-05-28 | 2013-05-28 | Vegetation-covered area high-spectrum remote-sensing altered mineral extracting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103383348B (en) |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10039244B2 (en) * | 2014-03-04 | 2018-08-07 | Greenonyx Ltd | Systems and methods for cultivating and distributing aquatic organisms |
| CN104077755B (en) * | 2014-07-10 | 2017-01-18 | 核工业北京地质研究院 | Remote sensing lithological information enhancement method based on chromatic aberration |
| CN104318550A (en) * | 2014-09-27 | 2015-01-28 | 励盼攀 | Eight-channel multi-spectral imaging data processing method |
| CN105068136A (en) * | 2015-07-27 | 2015-11-18 | 中国地质调查局武汉地质调查中心 | Potential positioning evaluation method for copper and gold mine of Indo-China peninsula demonstration zone based on multi-source information |
| CN105606537A (en) * | 2015-08-31 | 2016-05-25 | 山东科技大学 | Mineral type remote sensing recognition method based on multi-type spectral feature parameter collaboration |
| CN107291798A (en) * | 2017-05-09 | 2017-10-24 | 吉林大学 | A kind of extracting method and system of geochemistry data abnormal information |
| CN108171210B (en) * | 2018-01-18 | 2020-06-19 | 中国地质科学院矿产资源研究所 | Method and system for extracting remote sensing abnormal information of alteration of covered area of planting |
| CN108280433B (en) * | 2018-01-25 | 2020-06-09 | 中国地质科学院矿产资源研究所 | Identification method and system for gold mine under planting covered area |
| CN108427964B (en) * | 2018-03-05 | 2020-05-12 | 中国地质科学院矿产资源研究所 | Method and system for fusing remote sensing image and geochemistry |
| CN108593575A (en) * | 2018-04-18 | 2018-09-28 | 成都理工大学 | A kind of multi-spectrum remote sensing alteration information metallogenic prognosis method based on wave spectrum hourglass tool |
| CN109145881B (en) * | 2018-10-09 | 2021-06-22 | 中国地质科学院矿产资源研究所 | Remote sensing image paste salt information extraction method and device |
| CN109557031A (en) * | 2018-11-06 | 2019-04-02 | 核工业北京地质研究院 | A kind of rock core EO-1 hyperion Information extraction method |
| CN109961087B (en) * | 2019-02-01 | 2021-10-08 | 中国地质科学院矿产资源研究所 | Abnormal remote sensing information extraction method and device based on spatial data set analysis |
| CN110414738B (en) * | 2019-08-01 | 2022-03-22 | 吉林高分遥感应用研究院有限公司 | Crop yield prediction method and system |
| CN110706212A (en) * | 2019-09-10 | 2020-01-17 | 甘肃省地质调查院 | Mineralization alteration information extraction method, terminal equipment and storage medium |
| CN111199201B (en) * | 2020-03-06 | 2022-02-08 | 吉林大学 | General optimization extraction method for multispectral remote sensing mineralization alteration information |
| CN112381882A (en) * | 2020-11-04 | 2021-02-19 | 山东大学 | Unmanned aerial vehicle image automatic correction method carrying hyperspectral equipment |
| CN112577906B (en) * | 2020-11-18 | 2023-09-22 | 上海市园林科学规划研究院 | Urban green land soil moisture content detection method |
| CN113670836B (en) * | 2021-07-27 | 2024-03-12 | 中国地质调查局西安矿产资源调查中心 | Method and device for extracting hyperspectral remote sensing altered mineral information |
| CN114397251A (en) * | 2021-12-31 | 2022-04-26 | 核工业北京地质研究院 | Remote sensing identification method for uranium mineralization alteration information in planting covered area |
| CN114993965B (en) * | 2022-05-13 | 2023-04-18 | 中煤嘉沣(湖南)环保科技有限责任公司 | Automatic pollution source identification method and system |
| CN115753632B (en) * | 2022-10-19 | 2024-05-31 | 山东大学 | Method and system for identifying bad geologic bodies in tunnel in real time based on image spectrum |
| CN116754511B (en) * | 2023-08-18 | 2023-10-27 | 天津博霆光电技术有限公司 | Indocyanine green detection method and system based on spectrum technology |
| CN118090636A (en) * | 2024-04-25 | 2024-05-28 | 吉林大学 | Method for processing month observation data by foundation visible light hyperspectral imager |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6819798B2 (en) * | 2000-07-27 | 2004-11-16 | Bae Systems Information Electronic Systems Integration Inc. | Spectral drift and correction technique for hyperspectral imaging systems |
| CN101403795A (en) * | 2008-11-18 | 2009-04-08 | 北京交通大学 | Remote sensing survey method and system for estimating tree coverage percentage of city |
| CN101975941A (en) * | 2010-09-08 | 2011-02-16 | 吉林大学 | Portable remote sensing image information field acquisition system and acquisition method |
| CN102012528A (en) * | 2010-11-23 | 2011-04-13 | 北京理工大学 | Hyperspectral remote sensing oil-gas exploration method for vegetation sparse area |
| CN102426625A (en) * | 2011-09-21 | 2012-04-25 | 中国国土资源航空物探遥感中心 | Method for extracting porphyry copper typical altered zone mineral information based on ASTER satellite data |
-
2013
- 2013-05-28 CN CN201310204230.5A patent/CN103383348B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6819798B2 (en) * | 2000-07-27 | 2004-11-16 | Bae Systems Information Electronic Systems Integration Inc. | Spectral drift and correction technique for hyperspectral imaging systems |
| CN101403795A (en) * | 2008-11-18 | 2009-04-08 | 北京交通大学 | Remote sensing survey method and system for estimating tree coverage percentage of city |
| CN101975941A (en) * | 2010-09-08 | 2011-02-16 | 吉林大学 | Portable remote sensing image information field acquisition system and acquisition method |
| CN102012528A (en) * | 2010-11-23 | 2011-04-13 | 北京理工大学 | Hyperspectral remote sensing oil-gas exploration method for vegetation sparse area |
| CN102426625A (en) * | 2011-09-21 | 2012-04-25 | 中国国土资源航空物探遥感中心 | Method for extracting porphyry copper typical altered zone mineral information based on ASTER satellite data |
Non-Patent Citations (1)
| Title |
|---|
| 陈圣波等.植被覆盖区卫星高光谱遥感岩性分类.《吉林大学学报(地球科学版)》.2012,第42卷(第6期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103383348A (en) | 2013-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103383348B (en) | Vegetation-covered area high-spectrum remote-sensing altered mineral extracting method | |
| van der Meero et al. | Cross correlogram spectral matching: application to surface mineralogical mapping by using AVIRIS data from Cuprite, Nevada | |
| CN100547438C (en) | A kind of oil-gas exploration method and system | |
| Huang et al. | New optimized spectral indices for identifying and monitoring winter wheat diseases | |
| Zhang et al. | Comparison between wavelet spectral features and conventional spectral features in detecting yellow rust for winter wheat | |
| CN102507474B (en) | Method and system for identifying oil spilling target of ship | |
| CN102053083B (en) | Coal quality characteristic on-line measuring method based on partial least square method | |
| Goodwin et al. | Assessing plantation canopy condition from airborne imagery using spectral mixture analysis and fractional abundances | |
| Luo et al. | Detecting aphid density of winter wheat leaf using hyperspectral measurements | |
| Mateen et al. | The role of hyperspectral imaging: A literature review | |
| CN108195771A (en) | A kind of ocean target in hyperspectral remotely sensed image target identification method | |
| Ni et al. | Mineral identification and mapping by synthesis of hyperspectral VNIR/SWIR and multispectral TIR remotely sensed data with different classifiers | |
| CN105606537A (en) | Mineral type remote sensing recognition method based on multi-type spectral feature parameter collaboration | |
| Andrade et al. | Tropical soil order and suborder prediction combining optical and X-ray approaches | |
| Agar et al. | Remote sensing for mineral exploration–A decade perspective 1997-2007 | |
| CN109839356B (en) | Mineralization and alteration information extraction method based on WorldView-3 satellite data | |
| Kokaly | Spectroscopic remote sensing for material identification, vegetation characterization, and mapping | |
| Park et al. | Quantification of rock mass weathering using spectral imaging | |
| Kruse | Expert system analysis of hyperspectral data | |
| CN116797935A (en) | Rock and ore hyperspectral remote sensing image extraction method and device and computer equipment | |
| Burai et al. | Mapping aquatic vegetation of the Rakamaz-Tiszanagyfalui Nagy-morotva using hyperspectral imagery | |
| Yuan et al. | Selection of spectral channels for satellite sensors in monitoring yellow rust disease of winter wheat | |
| Cui et al. | Estimation of copper concentration of rocks using hyperspectral technology | |
| Biswas et al. | Discernment of complex lithologies utilizing different scattering and textural components of SAR and optical data through machine learning approaches in Jaisalmer, Rajasthan, India | |
| Oskouie et al. | A geostatistically based preprocessing algorithm for hyperspectral data analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190220 Address after: 130012 No. 352 Xinghuo Road, Chaoyang District, Changchun City, Jilin Province Patentee after: Jilin Jida Holdings Limited Address before: 130012 No. 2699 Qianjin Street, Jilin, Changchun Patentee before: Jilin University |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190311 Address after: Room 1532, Liwang Square, 996 Qianjin Street, Chaoyang District, Changchun City, Jilin Province, 130012 Patentee after: Jilin High-resolution Remote Sensing Applied Research Institute Co., Ltd. Address before: 130012 No. 352 Xinghuo Road, Chaoyang District, Changchun City, Jilin Province Patentee before: Jilin Jida Holdings Limited |
|
| TR01 | Transfer of patent right |