CN1924610A - Method for inversing forest canopy density and accumulating quantity using land satellite data - Google Patents
Method for inversing forest canopy density and accumulating quantity using land satellite data Download PDFInfo
- Publication number
- CN1924610A CN1924610A CN 200510093690 CN200510093690A CN1924610A CN 1924610 A CN1924610 A CN 1924610A CN 200510093690 CN200510093690 CN 200510093690 CN 200510093690 A CN200510093690 A CN 200510093690A CN 1924610 A CN1924610 A CN 1924610A
- Authority
- CN
- China
- Prior art keywords
- forest
- accumulation
- canopy density
- image
- inverting
- 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.)
- Pending
Links
Images
Landscapes
- Image Processing (AREA)
Abstract
This invention relates to one remote image technique field and especially to application technique method by use of LANDSAT TM data forest crown degree and storage, which comprises the following steps: a, pre-processing forest remote image to correct atmosphere and earth shape correction; then using statistical knowledge to compute land satellite data source each waveband brightness, regression plant index NDVI, changed brightness, green and moisture factors and forest crown degrees, storage relative parameter to find out parameters of the forest; finally finding out parameters and forest crown degree and storage relation equation.
Description
Technical field
The present invention relates to the remote sensing images technical field, particularly a kind of practical and technical methods that utilizes LANDSAT TM (U.S.'s Landsat special topic imaging drawing equipment) data inversion forest canopy density and accumulation.
Background technology
Because the forest land remote sensing images wave band DN value in area, mountain region is subjected to topographic relief, sun altitude, the meteorological condition influence is bigger.Therefore should carry out pre-service to remote sensing image to be analyzed, comprise geometric correction, atmosphere is corrected and landform is corrected.Classify to correcting later image then, the purpose of classification is that non-forest land area mask is fallen, and purpose is inverting canopy density, generation grade figure on the area, forest land.
1. domestic research level
Traditional forest canopy density is measured general by field measurement, and only can obtain the data on some points, is unfavorable for survey region or the space distribution and the variation of interior canopy density on a large scale.The development of remote sensing technology makes becomes possibility to forest canopy density estimation on a large scale.There is the scholar once this to be carried out research, the method that adopts is that the canopy density of determining with ground investigation are dependent variable, gray-scale value and gray level ratio with TM (Thematic Mapper thematic mapper) the some wave bands of data are independent variable, consider information such as the gradient, position, slope and dominant tree group simultaneously, adopt multivariate linear model estimation canopy density.Another kind method is from sample ground corresponding RS (Remote Sensing remote sensing) and GIS (GeographicInformation System) information, adopt the mountain range estimation principle, filter out the main information that influences the canopy density estimation with computer emulation method, set up equation between they and the canopy density, thereby realize canopy density are estimated.
2. the problem of Cun Zaiing
In fact because Effect of Environmental such as landform directly utilize the canopy density of six wave band data invertings of TM forest very difficult.
3. invention starting point
Finding out by statistical regression analysis influences the forest canopy density and the most significant factor of accumulation, sets up this factor and forest canopy density and accumulation equation of linear regression, and remote sensing images are carried out inverting.
Summary of the invention
The object of the present invention is to provide a kind of method of utilizing Landsat data inversion forest canopy density and accumulation.
Inventive concept
A kind of method of utilizing Landsat data inversion forest canopy density and accumulation,
At first the forest land remote sensing images are carried out pre-service, promptly atmosphere is corrected and the landform correction;
Utilize the knowledge of statistical mathematics then, obtain each wave band brightness value of Landsat data source, normalized differential vegetation index NDVI, the related coefficient of the factor such as the brightness after the red-tasselled official hat conversion, green degree and humidity and forest land canopy density, accumulation is found out forest land canopy density, the strongest parameter of accumulation influence;
Find out the correlativity equation of parameter and forest land canopy density, accumulation at last,, and generate grade figure by correlativity equation inverting forest land canopy density, accumulation.
Set up equation by statistical regression methods, remote sensing images are carried out canopy density and the accumulation of inverting to obtain forest, and carry out remote sensing classification estimation.
Atmosphere is corrected
The thin cloud of L (beam is along the woods algorithm) method-removal:
Carry out the thin cloud removal method of a kind of atmosphere that atmosphere is corrected according to beam along thought oneself programming of woods algorithm.Verify the feasibility of this algorithm, the solution of simultaneously this algorithm being used practical problems in 863 projects comes up.
Principle is, because visible light wave range is subjected to atmospheric effect bigger, and 4,5,7 wave band wavelength are longer, be subjected to atmospheric effect very little, therefore 1,4 wave bands are done the influence that ratio can highlight Bao Yun, this ratio image is cut apart to be told affected zone and unaffected zone more accurately.Utilize 4,5,7 these three images synthetic to carry out the K-mean cluster, can objectively carry out type in theory and divide real surface than long-wave band.Suppose in the same scape TM image or in the same survey region, the reflection case of each face of land type is identical, like this since, just can be in each cluster, replace the reflectivity of fuzzy region with the average reflectance of clear area, thereby make the influence of eliminating thin cloud in the image as a result largely.
Designed three functions and finished the work,
(1) RatioClu (TM image file name is cut apart number, cycle index, threshold value, segmentation result filename) finds clear and the classification number fuzzy region correspondence from split image;
(2) Kmeans (TM image file name, clusters number, cycle index, threshold value, cluster result filename);
(3) AtmCorect (segmentation result filename, cluster result filename, the clear number of cutting apart, the fuzzy number of cutting apart, clusters number, TM image file name, image acquisition time, incident angle, albedo image filename).
Cycle index in function (1) and (2) all is 200, and threshold value is 0.01.Concrete job step and result are as shown in Figure 3.
6s atmosphere correcting method:
Because the polytrope of the reflectivity that diversity caused of face of land object makes that it is very difficult isolating the path radiation term from the radiation value of each pixel of remote sensing images, so just must choose and make the surface radiation value can ignore or have radiation but very little atural object but also the atural object of estimation ground surface reflectance that can be more accurate if we will obtain gasoloid information with radiation term.The blue light of dense vegetation in a large number such as deciduous forest and Tropical forests etc. and infrared reflectivity are all very low.Therefore, many places with forest or the dense vegetation of bulk can obtain gasoloid information with this method of DDV (Dark denseVegetation), thereby carry out atmospheric correction.
For bearing calibration,, therefore intend adopting dense vegetation method (DDV) to obtain gasoloid information, and then utilize the 6S model that image is carried out atmospheric correction because main processing is the wood land.This module is mainly divided 4 parts: (1) utilizes the apparent reflectance of aeropause under the different solar zenith angles of 6S Model Calculation, ground surface reflectance and the aerosol optical depth condition, sets up look-up table (LUT); (2) utilize the middle infrared channel of TM from the dark target in TM image recognition ground (dense vegetation or water body); (3) utilize LUT to pass through the method inverting aerosol optical depth of Lagrange linear interpolation; (4) aerosol optical depth that utilizes inverting to obtain obtains the atmospheric correction parameter by the 6S Model Calculation again, and entire image is carried out atmospheric correction.
Process flow diagram: (Fig. 4)
We are according to Kaufman and Sendra (1988) method, and from TM image calculation atmospheric aerosol optical thickness, the measured value that provides by spectral band is to dark target inverting unit carrying out one by one based on the 6S model.Inversion method mainly is based on following three hypothesis:
A) have dense vegetation, its reflectivity satisfies following relation:
ρ′
0.47＝ρ
* 2.1/4，ρ′
0.66＝ρ
* 2.1/2。
B) gas characteristic of atmosphere is known.
C) the gasoloid pattern of atmosphere is known.
After above three hypothesis all satisfy, just can be from the land aerosol optical depth in inverting clear sky zone on the image, and carry out atmospheric correction, the process of its method is as follows:
(1) chooses dark picture dot, determine face of land reflectivity, the hypothesis that it is constant that Kaufman once proposed dense vegetation spectral reflectivity, but on entire image, it is imprecise that the hypothesis that dense vegetation spectral reflectivity is a constant is used the area that we do not know, we (consider and satisfy the ρ ' that Kaufman proposes according to the corresponding relation of 2.1 μ m reflectivity and 0.47 μ m, 0.66 μ m visible light wave range reflectivity at last
0.47=0.01 (ρ represents reflectivity), ρ '
0.66=0.02 hypothesis is got 0.036＜ρ
* 2.1＜0.044), determines that dark target earth surface reflection rate is: ρ '
0.47=ρ
* 2.1/ 4, ρ '
0.66=ρ
* 2.1/ 2;
(2) select the gasoloid pattern according to the weather of locality and the state of ground of meteorological data and surrounding area;
(3), utilize the aerosol optical depth of the apparent reflectance calculating 550nm of 1,3 wave bands respectively to each dark picture dot;
(4) aerosol optical depth of final 550nm is got the mean value of the optical thickness that 1,3 wave band calculates;
(5) the gasoloid thickness that inverting is obtained is brought the 6S Model Calculation into and is obtained the atmospheric correction coefficient, and entire image is proofreaied and correct.
Operation instruction:
1. import the radiation gain coefficient of each wave band
(obtain manner: the GAINS parameter value of the 1st of GAINS/BIASES the, 2,3,4,5,7 value centerings in the raw video header file)
2. import the radiation offset coefficient of each wave band
(obtain manner: the BIASES parameter value of the 1st of GAINS/BIASES the, 2,3,4,5,7 value centerings in the raw video header file)
3. the user selects which wave band of remedial frames (obtain manner: the user must choose the 1st, 3,4,7 wave bands, and other is any)
4. program needs to calculate NDVI (Normal Difference Vegetationindex) value in operational process, uses the 3rd, 5. wave band; And the apparent brightness (ApparentRellectance) of the 7th wave band (Mir) judges dark object run, finds the apparent brightness of first wave band of dark target correspondence afterwards by the position of the dark target that finds, by concerning ρ '
0.47=ρ
* 2.1/ 4 utilize dark target to try to achieve the actual ground emissivity of this dark object element at first wave band at the reflectivity of the 7th wave band, provide interpolation parameter for making an inventory table, calculate gasoloid thickness)
5. be entered as picture date and solar zenith angle
(obtain manner: ACQUISITION DATE parameter value in the raw video header file and solar zenith angle [90 degree-SUN ELEVATION] parameter value)
6. select atmospherical model (obtain manner: the user selects, middle latitude commonly used summer, middle latitude winter) from the 6s subordinate list
(note: atmospherical model and gasoloid pattern selected in the 6s correction procedure are made an inventory the selected pattern of table correspondence one by one with setting up.)
7. (note: atmospherical model and gasoloid pattern selected in the 6s correction procedure are made an inventory the selected pattern of table correspondence one by one with setting up to select gasoloid pattern (obtain manner: the user selects from the 6s subordinate list, uses the continent always).)
8. the height above sea level average height of input picture (obtain manner: the user specifies, often from the topomap estimation)
9. specify the making an inventory of table lut1 and red (Red) wave band of making an inventory of blue (blue) wave band to show lut3 (obtain manner: the self-built or use acquiescence default (summers in winter two cover is arranged) of user respectively
10. point to the TM file (obtain manner: the user specifies) that will carry out radiation correcting
Correct back output file (obtain manner: the user specifies) 11. point to
6s parameter subordinate list:
| Parametric variable and implication | Input value | Remarks | |
| The igeom geometric condition | ?Default | Month, day; (imaging date) sza, (solar zenith angle) | Angular unit degree of being |
| The idatm atmospherical model | ?＝0 | No gas absorption | |
| ?＝1 | The torrid zone | ||
| ?＝2 | Middle latitude summer | ||
| ?＝3 | Middle latitude winter | ||
| ?＝4 | Subarctiv summer | ||
| ?＝5 | Subarctiv winter | ||
| ?＝6 | United States standard atmosphere 62 | ||
| ?＝7 | The user imports profile: highly (km), air pressure (mb), temperature (k), H 2O density (g/m 3)，O 3Density (g/m 3) be P, T, UH2O, UO3 | 34 layers of sounding data, every layer of 5 parameter, totally 34 * 5 data | |
| ?＝8 | Moisture content (g/cm 2), ozone content (cm-atm) is U_H2O, U_O3; | Other use United States standard atmosphere | |
| The iaer aerosol type | ?＝0 | No gasoloid | |
| ?＝1 | The continent | ||
| ?＝2 | The ocean | ||
| ?＝3 | The city | ||
| ?＝4 | Self-defined: dust, water-soluble, the ocean, coal smoke is Dust_Like, Oceanic, Water_Soluble, Soot; | Four kinds of percentages of ingredients | |
| ?＝5 | The desert | ||
| ?＝6 | Biomass combustion | ||
| ?＝7 | The stratosphere pattern | ||
| Xps target height above sea level (km) | ?≥0 | Absolute value is the target sea level elevation | Unit (kilometer) |
Dark target subraction
Atmospheric condition is constantly to change on space-time, and it is comparatively all even stable that atmospheric molecule distributes at time space, and aerocolloidal change in time and space is violent, and it is very difficult to estimate to get up.Atmosphere has absorption characteristic selectively to different POPs, thereby also can produce different influences to the image of the different-waveband of TM image.Atmosphere is mainly by optionally absorbing and the energy of scattering from the different POPs of the sun.At infrared spectral coverage, the effect of atmosphere mainly shows steam, and carbon dioxide etc. are to the absorption of energy, and atmospheric molecule and aerocolloidal scattering can be ignored.In visible spectrum, the effect of atmosphere shows as the scattering process to energy highlightedly, the Mie scattering that Rayleigh scattering that produces comprising atmospheric molecule and gasoloid cause, and the absorption of atmosphere is very faint then.The TM image is corrected exactly in order to eliminate the influence to atmosphere of atmospheric scattering and absorption.
Suppose near infrared and on the section of infrared POP the TM image be not subjected to the influence of atmospheric aerosol scattering, opposite visible light wave range is along with the POP energy that shortening of wavelength is scattered is strengthened gradually.We carry out atmosphere based on following hypothesis thought to image and correct.
The atmospheric condition of a whole scape TM image is a homogeneous; In the TM/ETM+ satellite image we can to find reflectivity from the image of each wave band on the section of visible light POP respectively be zero or very near zero picture dot, for example: pure deep water, the shade of atural object, therefore theoretically like this reflectivity be zero or should be zero in the DN of image value very near zero dark pixel, and in fact since atmospheric aerosol scattering to influence its DN value big zero, we just can think that such result is the result of the effect of sunshine and atmospheric aerosol scattering, so the operation that the influence of in the reality satellite image being removed atmospheric aerosol scattering can be expressed as right DN value promptly: DN
Result=DN
Raw-DN
Dark
Parameter declaration:
Constant in the formula: λ wavelength
TM Esunl TAUZ Gain
λRadiation gain,
Band Bias
λRadiation offset,
TM1 195.29 0.70 DN
λBe TM gradation of image value,
TM2 182.74 0.78 L
λ sensorApparent spoke brightness value,
L
λ maxMaximum spectral radiance value (255),
TM3 155.00 0.85 L
λ minMinimum spectrum spoke brightness value (0),
TM4 104.08 0.91 L
λ rangeThe spoke brightness range (0-255) of TM data,
TM5 22.075 * * R
λReflectivity (earth surface reflection rate and apparent reflectance),
TM7 7.496 * * L
λ hazeThe journey radiation,
Esun
λThe outer solar illumination of atmosphere,
The θ solar zenith angle
The D solar distance,
TAU
VThe atmospheric transmittance of sensor is arrived on the face of land,
TAU
ZThe sun is saturating to the atmosphere on the face of land
Operation instruction:
The radiation gain coefficient of 1 each wave band of input
(obtain manner: the GAINS parameter value of the 1st of GAINS/BIASES the, 2,3,4,5,7 value centerings in the raw video header file)
The radiation offset coefficient of 2 each wave bands of input
(obtain manner: the BIASES parameter value of the 1st of GAINS/BIASES the, 2,3,4,5,7 value centerings in the raw video header file)
3 input sun altitude (obtain manners: the SUN ELEVATION parameter value in the raw video header file)
4 users select which wave band of remedial frames (obtain manner: the user specifies)
5 point to the TM file (obtain manner: the user specifies) that will carry out radiation correcting
6 point to correction back output file (obtain manner: the user specifies)
The invention process of canopy density and accumulation inversion method:
1. at first with ground investigation sample ground information input computing machine,, make bit map/bitmap and change image processing system over to, on the image that is added to, find the corresponding wave band DN value of attribute information (canopy density) according to geographic coordinate then according to the geographic coordinate that GPS measures.
2. with the DN value of landsat image wave band 1,2,3,4,5 and 7, normalized differential vegetation index NDVI, accumulation and the canopy density of the factor such as the brightness after the red-tasselled official hat conversion, green degree and humidity and forest are carried out regretional analysis.
3. finding out accumulation and the strongest amount of canopy density correlativity with forest, is respectively that the humidity factor and the forest canopy density correlativity of red-tasselled official hat conversion is the strongest, and the accumulation of wave band 1 and forest is the strongest.
4. set up the regression equation in advance of humidity factor and forest canopy density, the power function equation of wave band 1 and forest reserves.
5. set up equation above utilizing remote sensing images are carried out inverting, obtain forest land canopy density and accumulation.
6. to carrying out the classification estimation with forest land canopy density and the accumulation obtained.
Process flow diagram: (Fig. 5)
Specific implementation
Canopy density:
Through after the regretional analysis, discovery is the strongest with the canopy density correlativity that landsat image carries out the humidity factor of red-tasselled official hat conversion (Tasseled Cap) and forest, utilize the equation of linear regression of this foundation that remote sensing images are carried out inverting to obtain the canopy density of forest, carry out remote sensing classification estimation at last.
Accumulation:
The analysis found that, find the data of landsat image wave band 1 are relevant with forest reserves the strongest, utilize this to set up power function equation remote sensing images are carried out inverting to obtain the accumulation of forest, carry out remote sensing classification estimation at last.
Description of drawings
Fig. 1 is the process flow diagram that the present invention utilizes Landsat data inversion forest canopy density and accumulation method.
Fig. 2 is the L method flow diagram that the present invention utilizes Landsat data inversion forest canopy density and accumulation method.
Fig. 3 is the image of L method atmosphere of the present invention each process when correcting.
Fig. 4 is a 6s method flow diagram of the present invention.
Fig. 5 is the process flow diagram of forest canopy density of the present invention, accumulation inversion method.
Fig. 6 is forest land canopy density grade figure.
Fig. 7 is forest land accumulation grade figure.
Embodiment
The method of utilizing Landsat data inversion forest canopy density and accumulation of Fig. 1, its step is as follows:
1) Landsat TM image being carried out atmosphere corrects;
2) the TM image of atmosphere being corrected carries out landform to be corrected;
3) will correct good image and classify, after the classification non-forest land part mask be fallen;
4) statistical regression model that utilizes training area to set up carries out inverting to the pretreated image of process, obtains canopy density and accumulation;
5) the grade figure of output canopy density and accumulation.
Atmosphere is corrected
L method-removal approaches cloud: (Fig. 2)
(1) selects fuzzy region; (band 1/band 4)
(2) adopt clustering method that near-infrared band (4,5,7 wave band) is distinguished different cover types;
(3) in identical cluster, represent the value of fuzzy pixel with the mean value of clear pixel.
Choose a part that influenced by Bao Yun in the northeast Heilongjiang river one scape TM image, adopt above-mentioned algorithm and routine processes result as follows.
Fig. 3 has provided the image of each process when using L method atmosphere to correct, wherein:
Figure a is that image is through wave band 1,2 and 3 synthetic striographs.
Figure b is the ratio figure of image band 1 and wave band 4.
Figure c is for utilizing wave band 4,5 and 7 classify (totally 10 classes).
Figure d is process wave band 1,2 and 3 synthetic striographs after correcting through L method atmosphere.
Interpretation of result: from Fig. 3 (a) and 4 classes of (b) utilizing the ratio image segmentation to come out as can be seen, clear area in the corresponding band image of blue 1 (number of cutting apart is 3), fuzzy region in the corresponding band image of yellow 2 (number of cutting apart is 4), these 2 parameters are as the input parameter of function (3).The reflectivity in river is very low, can be branched away accurately on the ratio image.Green 3 parts are the higher bare areas of actual face of land reflectivity among the figure, although they also are subjected to the influence of mist, but utilize this algorithm to be divided into an other class, to not carry out the conversion process of reflectivity in follow-up correction, therefore being segmented in automatically with machine still has certain weak point in this algorithm.Fig. 3 (c) is the triband cluster result, and designing user can be imported the cycle index of cluster numbers, K-mean algorithm, the threshold value of cluster centre variable quantity in the program.The cluster numbers here is 10 classes, can represent the typical feature type of survey region.Fig. 3 (d) is process wave band 1,2 and the 3 synthetic striographs after correcting through L method atmosphere.
6s (the Second Simulation of the Satellite Signal in the Solar Spectrum) atmosphere correcting method: (Fig. 4)
(1) utilizes the apparent reflectance of aeropause under the different solar zenith angles of 6S Model Calculation, ground surface reflectance and the aerosol optical depth condition, set up look-up table;
(2) utilize the middle infrared channel of TM from the dark target in TM image recognition ground;
(3) utilize the method inverting aerosol optical depth of LUT (lool up table makes an inventory table) by the Lagrange linear interpolation;
(4) aerosol optical depth that utilizes inverting to obtain obtains the atmospheric correction parameter by the 6S Model Calculation again, and entire image is carried out atmospheric correction.
Dark target deducts method:
1. import a wave band in the TM image;
2. search the DN value of the pixel of brightness minimum;
3. atmosphere is corrected DNresult=DNraw-Dndark;
4. by formula one the DN value is converted into radiance;
1. transfer spoke brightness to reflectivity by formula two, three.
The process of canopy density and accumulation inversion method:
Process flow diagram: (Fig. 5)
Forest land canopy density, accumulation inverting
(1) at first with ground investigation sample ground information input computing machine, geographic coordinate according to GPS mensuration, make bit map/bitmap and change image processing system over to, on the image that is added to, find corresponding wave band DN (the digital number gray-scale value) value of attribute information (canopy density) according to geographic coordinate then;
(2) with the DN value of landsat image wave band 1,2,3,4,5 and 7, normalized differential vegetation index NDVI, accumulation and the canopy density of the factor such as the brightness after the red-tasselled official hat conversion, green degree and humidity and forest are carried out regretional analysis; Finding out accumulation and the strongest amount of canopy density correlativity with forest, is respectively that the humidity factor and the forest canopy density correlativity of red-tasselled official hat conversion is the strongest, and the accumulation of wave band 1 and forest is the strongest;
(3) set up the in advance regression equation of humidity factor and forest canopy density, the power function equation of wave band 1 and forest reserves;
(4) set up equation above the utilization remote sensing images are carried out inverting, obtain forest land canopy density and accumulation;
(5) to carrying out the classification estimation with forest land canopy density and the accumulation obtained.
Specific implementation
Canopy density:
Through after the regretional analysis, discovery is the strongest with the canopy density correlativity that landsat image carries out the humidity factor of red-tasselled official hat conversion (Tasseled Cap) and forest, utilize the equation of linear regression of this foundation that remote sensing images are carried out inverting to obtain the canopy density of forest, carry out remote sensing classification estimation at last.
Accumulation:
The analysis found that, find the data of landsat image wave band 1 are relevant with forest reserves the strongest, utilize this to set up power function equation remote sensing images are carried out inverting to obtain the accumulation of forest, carry out remote sensing classification estimation at last.
Inversion result figure (Fig. 6,7)
Utilize the inversion equation of setting up, generate grade figure carry out computing through pretreated image.Fig. 6 is for generating forest land canopy density grade figure after the inverting, and wherein black 4 is represented the non-forest land area that mask falls, and green 3 represents the lower area of forest land canopy density, and blue 1 represents the medium area of forest land canopy density, and redness 5 is represented the higher area of forest land canopy density.
Utilize the inversion equation of setting up, generate grade figure carry out computing through pretreated image.Fig. 7 is for generating forest land accumulation grade figure after the inverting, and wherein black 4 is represented the non-forest land area that mask falls, and green 3 represents the lower area of forest land canopy density, and blue 1 represents the medium area of forest land canopy density, and redness 5 is represented the higher area of forest land canopy density.
Claims (7)
1. method of utilizing Landsat data inversion forest canopy density and accumulation, its feature
Be, at first the forest land remote sensing images carried out pre-service, promptly atmosphere is corrected and the landform correction;
Utilize the knowledge of statistical mathematics then, obtain each wave band brightness value of Landsat data source, normalized differential vegetation index NDVI, the related coefficient of the factor such as the brightness after the red-tasselled official hat conversion, green degree and humidity and forest land canopy density, accumulation is found out forest land canopy density, the strongest parameter of accumulation influence;
Find out the correlativity equation of parameter and forest land canopy density, accumulation at last,, and generate grade figure by correlativity equation inverting forest land canopy density, accumulation.
2. according to the method for utilizing Landsat data inversion forest canopy density and accumulation of claim 1, its concrete steps are as follows:
1) Landsat TM image being carried out atmosphere corrects;
2) the TM image of atmosphere being corrected carries out landform to be corrected;
3) will correct good image and classify, after the classification non-forest land part mask be fallen;
4) statistical regression model that utilizes training area to set up carries out inverting to the pretreated image of process, obtains canopy density and accumulation;
5) the grade figure of output canopy density and accumulation.
3. according to the method for utilizing Landsat data inversion forest canopy density and accumulation of claim 1, it is characterized in that atmosphere is corrected
L method-removal approaches cloud:
(1) selects fuzzy region;
(2) adopt clustering method that near-infrared band is distinguished different cover types;
(3) in identical cluster, represent the value of fuzzy pixel with the mean value of clear pixel.
4. according to the method for utilizing Landsat data inversion forest canopy density and accumulation of claim 3, it is characterized in that atmosphere is corrected
(1) chooses dark picture dot, determine face of land reflectivity, on entire image, it is imprecise that the hypothesis that dense vegetation spectral reflectivity is a constant is used the area of not known, at last, determine that dark target earth surface reflection rate is: ρ ' according to the corresponding relation of 2.1 μ m reflectivity and 0.47 μ m, 0.66 μ m visible light wave range reflectivity
0.47=ρ
* 2.1/ 4, ρ '
0.66=ρ
* 2.1/ 2;
(2) select the gasoloid pattern according to the weather of locality and the state of ground of meteorological data and surrounding area;
(3), utilize the aerosol optical depth of the apparent reflectance calculating 550nm of 1,3 wave bands respectively to each dark picture dot;
(4) aerosol optical depth of final 550nm is got the mean value of the optical thickness that 1,3 wave band calculates;
(5) the gasoloid thickness that inverting is obtained is brought the 6S Model Calculation into and is obtained the atmospheric correction coefficient, and entire image is proofreaied and correct.
5. according to the method for utilizing Landsat data inversion forest canopy density and accumulation of claim 1, it is characterized in that 6s atmosphere correcting method:
(1) utilizes the apparent reflectance of aeropause under the different solar zenith angles of 6S Model Calculation, ground surface reflectance and the aerosol optical depth condition, set up look-up table;
(2) utilize the middle infrared channel of TM from the dark target in TM image recognition ground;
(3) utilize LUT to pass through the method inverting aerosol optical depth of Lagrange linear interpolation;
(4) aerosol optical depth that utilizes inverting to obtain obtains the atmospheric correction parameter by the 6S Model Calculation again, and entire image is carried out atmospheric correction.
6. according to the method for utilizing Landsat data inversion forest canopy density and accumulation of claim 1, it is characterized in that forest land canopy density, accumulation inverting
(1) at first with ground investigation sample ground information input computing machine,, makes bit map/bitmap and change image processing system over to, on the image that is added to, find the wave band DN value of attribute information correspondence then according to geographic coordinate according to the geographic coordinate that GPS measures;
(2) with the DN value of landsat image wave band 1,2,3,4,5 and 7, normalized differential vegetation index NDVI, accumulation and the canopy density of the factor such as the brightness after the red-tasselled official hat conversion, green degree and humidity and forest are carried out regretional analysis; Finding out accumulation and the strongest amount of canopy density correlativity with forest, is respectively that the humidity factor and the forest canopy density correlativity of red-tasselled official hat conversion is the strongest, and the accumulation of wave band 1 and forest is the strongest;
(3) set up the in advance regression equation of humidity factor and forest canopy density, the power function equation of wave band 1 and forest reserves;
(4) set up equation above the utilization remote sensing images are carried out inverting, obtain forest land canopy density and accumulation;
(5) to carrying out the classification estimation with forest land canopy density and the accumulation obtained.
7. according to the method for utilizing Landsat data inversion forest canopy density and accumulation of claim 1, it is characterized in that,
Set up equation by statistical regression methods, remote sensing images are carried out canopy density and the accumulation of inverting to obtain forest, and carry out remote sensing classification estimation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510093690 CN1924610A (en) | 2005-09-01 | 2005-09-01 | Method for inversing forest canopy density and accumulating quantity using land satellite data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510093690 CN1924610A (en) | 2005-09-01 | 2005-09-01 | Method for inversing forest canopy density and accumulating quantity using land satellite data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1924610A true CN1924610A (en) | 2007-03-07 |
Family
ID=37817330
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200510093690 Pending CN1924610A (en) | 2005-09-01 | 2005-09-01 | Method for inversing forest canopy density and accumulating quantity using land satellite data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1924610A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101936777A (en) * | 2010-07-30 | 2011-01-05 | 南京信息工程大学 | Method for inversing air temperature of surface layer based on thermal infrared remote sensing |
| CN101315424B (en) * | 2008-07-29 | 2011-06-08 | 中国科学院对地观测与数字地球科学中心 | Multi-satellite remote sensing data integrated parallel ground pretreatment system |
| CN102162844A (en) * | 2010-12-07 | 2011-08-24 | 北京理工大学 | Rapid analogue method of synthetic aperture radar (SAR) large range forest scene remote sensing data |
| CN102231190A (en) * | 2011-07-08 | 2011-11-02 | 中铁第四勘察设计院集团有限公司 | Automatic extraction method for alluvial-proluvial fan information |
| CN102542276A (en) * | 2011-12-27 | 2012-07-04 | 中国科学院东北地理与农业生态研究所 | Method for rapidly extracting forest canopy density by applying Photoshop and Matlab |
| CN102667816A (en) * | 2009-12-22 | 2012-09-12 | 韦尔豪泽Nr公司 | Method and apparatus for predicting information about trees in images |
| CN103364309A (en) * | 2013-07-19 | 2013-10-23 | 中国医学科学院药用植物研究所 | Remote sensing monitoring method for wild rhubarb population structure and reasonable excavation amount |
| CN105403876A (en) * | 2015-12-24 | 2016-03-16 | 中国林业科学研究院资源信息研究所 | Measuring method of forest canopy density and device |
| CN106017367A (en) * | 2013-04-28 | 2016-10-12 | 中国林业科学研究院资源信息研究所 | Canopy density determining method and apparatus |
| CN107831501A (en) * | 2017-10-27 | 2018-03-23 | 北京林业大学 | A kind of method of ground laser radar simulation angle gauge measure Stand Volume |
| CN108151719A (en) * | 2017-12-07 | 2018-06-12 | 福州大学 | A kind of method for verifying topographic shadowing calibration result |
| CN108344400A (en) * | 2017-01-22 | 2018-07-31 | 北京林业大学 | A kind of horizontal normal case photography imagery measures the technical method of Forest Disaster state distribution |
| CN109785569A (en) * | 2019-01-28 | 2019-05-21 | 中科光启空间信息技术有限公司 | A kind of forest fire monitoring method based on 3S technology |
| CN112649372A (en) * | 2020-11-27 | 2021-04-13 | 中国科学院东北地理与农业生态研究所 | Method for inverting forest canopy density by remote sensing based on machine learning |
-
2005
- 2005-09-01 CN CN 200510093690 patent/CN1924610A/en active Pending
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101315424B (en) * | 2008-07-29 | 2011-06-08 | 中国科学院对地观测与数字地球科学中心 | Multi-satellite remote sensing data integrated parallel ground pretreatment system |
| CN102667816A (en) * | 2009-12-22 | 2012-09-12 | 韦尔豪泽Nr公司 | Method and apparatus for predicting information about trees in images |
| CN101936777A (en) * | 2010-07-30 | 2011-01-05 | 南京信息工程大学 | Method for inversing air temperature of surface layer based on thermal infrared remote sensing |
| CN102162844A (en) * | 2010-12-07 | 2011-08-24 | 北京理工大学 | Rapid analogue method of synthetic aperture radar (SAR) large range forest scene remote sensing data |
| CN102231190A (en) * | 2011-07-08 | 2011-11-02 | 中铁第四勘察设计院集团有限公司 | Automatic extraction method for alluvial-proluvial fan information |
| CN102231190B (en) * | 2011-07-08 | 2012-10-31 | 中铁第四勘察设计院集团有限公司 | Automatic extraction method for alluvial-proluvial fan information |
| CN102542276A (en) * | 2011-12-27 | 2012-07-04 | 中国科学院东北地理与农业生态研究所 | Method for rapidly extracting forest canopy density by applying Photoshop and Matlab |
| CN106017367A (en) * | 2013-04-28 | 2016-10-12 | 中国林业科学研究院资源信息研究所 | Canopy density determining method and apparatus |
| CN106017367B (en) * | 2013-04-28 | 2018-03-30 | 中国林业科学研究院资源信息研究所 | The assay method and device of a kind of canopy density |
| CN104121850B (en) * | 2013-04-28 | 2017-02-01 | 中国林业科学研究院资源信息研究所 | Canopy density measurement method and device |
| CN103364309B (en) * | 2013-07-19 | 2016-02-10 | 中国医学科学院药用植物研究所 | The remote-sensing monitoring method of Wild Rhubarb population structure and the rationally amount of excavating |
| CN103364309A (en) * | 2013-07-19 | 2013-10-23 | 中国医学科学院药用植物研究所 | Remote sensing monitoring method for wild rhubarb population structure and reasonable excavation amount |
| CN105403876A (en) * | 2015-12-24 | 2016-03-16 | 中国林业科学研究院资源信息研究所 | Measuring method of forest canopy density and device |
| CN105403876B (en) * | 2015-12-24 | 2018-01-30 | 中国林业科学研究院资源信息研究所 | The measuring method and device of forest canopy density |
| CN108344400A (en) * | 2017-01-22 | 2018-07-31 | 北京林业大学 | A kind of horizontal normal case photography imagery measures the technical method of Forest Disaster state distribution |
| CN107831501A (en) * | 2017-10-27 | 2018-03-23 | 北京林业大学 | A kind of method of ground laser radar simulation angle gauge measure Stand Volume |
| CN108151719A (en) * | 2017-12-07 | 2018-06-12 | 福州大学 | A kind of method for verifying topographic shadowing calibration result |
| CN109785569A (en) * | 2019-01-28 | 2019-05-21 | 中科光启空间信息技术有限公司 | A kind of forest fire monitoring method based on 3S technology |
| CN112649372A (en) * | 2020-11-27 | 2021-04-13 | 中国科学院东北地理与农业生态研究所 | Method for inverting forest canopy density by remote sensing based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1924610A (en) | Method for inversing forest canopy density and accumulating quantity using land satellite data | |
| US20190130547A1 (en) | Atmospheric compensation in satellite imagery | |
| CN108253943B (en) | Integrated monitoring method for enteromorpha in red tide based on satellite remote sensing image | |
| CN1710379A (en) | Atmosphere correction method of airosol optical thickness of aeronautical high-spectrum remote-sensing inversion boundary layer | |
| CN110645961B (en) | Forest resource dynamic change detection method based on remote sensing and NDVI | |
| CN102539336A (en) | Method and system for estimating inhalable particles based on HJ-1 satellite | |
| CN114819150B (en) | Remote sensing inversion method for primary productivity of polar region ocean in winter | |
| CN116519557B (en) | Aerosol optical thickness inversion method | |
| CN109253976B (en) | High-spectrum real-time radiometric calibration method based on light sensing module | |
| Tagle Casapia | Study of radiometric variations in Unmanned Aerial Vehicle remote sensing imagery for vegetation mapping | |
| CN108256186A (en) | A kind of pixel-by-pixel atmospheric correction method in line computation look-up table | |
| CN109543654A (en) | A kind of construction method for the modified vegetation index reflecting crop growth situation | |
| CN111198162B (en) | Remote sensing inversion method for urban surface reflectivity | |
| CN115856925A (en) | Multispectral remote sensing image water depth inversion method, medium and equipment based on chart data | |
| Lazuardi et al. | Remote sensing for coral reef and seagrass cover mapping to support coastal management of small islands | |
| CN108009392A (en) | A kind of the Remote Sensing Reflectance model construction and Analysis method of dense vegetation earth's surface | |
| Weisz et al. | International MODIS and AIRS processing package: AIRS products and applications | |
| CN112329790B (en) | Quick extraction method for urban impervious surface information | |
| CN103777205B (en) | Based on the self adaptive imaging method of the polynary parameter calibration model of remote sensing image DN value | |
| Fisher et al. | A long-term record of photosynthetically available radiation (PAR) and total solar energy at 38.6 N, 78.2 W | |
| CN116822141A (en) | Method for inverting optical thickness of night atmospheric aerosol by utilizing satellite micro-optic remote sensing | |
| CN106650673A (en) | Urban mapping method and device | |
| CN112964643B (en) | Method for correcting landform falling shadow of visible light wave band of remote sensing image | |
| CN101034476A (en) | Method for generating underwater non-shadow sonar remote sensing orthographic digital image by computer | |
| CN1755392A (en) | Extract the MODIS time series data synthetic method and the device thereof of fire-slass area |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |