CN105865424A - Nonlinear model-based multispectral remote sensing water depth inversion method and apparatus thereof - Google Patents
Nonlinear model-based multispectral remote sensing water depth inversion method and apparatus thereof Download PDFInfo
- Publication number
- CN105865424A CN105865424A CN201610228558.4A CN201610228558A CN105865424A CN 105865424 A CN105865424 A CN 105865424A CN 201610228558 A CN201610228558 A CN 201610228558A CN 105865424 A CN105865424 A CN 105865424A
- Authority
- CN
- China
- Prior art keywords
- water
- water depth
- depth value
- reflectivity
- waters
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C13/00—Surveying specially adapted to open water, e.g. sea, lake, river or canal
- G01C13/008—Surveying specially adapted to open water, e.g. sea, lake, river or canal measuring depth of open water
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Hydrology & Water Resources (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a nonlinear model-based multispectral remote sensing water depth inversion method and an apparatus thereof. The method comprises the following steps: acquiring the multispectral remote sensing image of a preset area and the actually measured control point water depth of a preset water area, and preprocessing the multispectral remote sensing image to obtain a preset area reflectivity; carrying out water-land separation on the preset area reflectivity through a near infrared waveband spectrum characteristic-based threshold technique to obtain the reflectivity of the water surface of the preset water area; establishing a nonlinear inversion model corresponding to the preset water area according to the water surface reflectivity and the actually measured control point water depth; and regressing the nonlinear inversion model through a stepwise regression algorithm, and inversing according to the regressed nonlinear inversion model to obtain the water depth of the preset water area. The highly-precise water depth of the island reef water area far from the land is rapidly obtained according to the nonlinear inversion model on the basis of the actually measured water depth, model establishing and the solving process are simple, and the nonlinear inversion model is suitable for various types of water depth inversion engineering and has good portability.
Description
Technical field
The present invention relates to application of satellitic remote sensing technical field, in particular to a kind of how light based on nonlinear model
Spectrum remote sensing inversion method and device.
Background technology
At present, along with the extensive application of satellite remote sensing technology, promoted the improvement of water-depth measurement method, wherein passed through remote sensing
The method of Depth extraction sounds the depth of the water becomes an advanced water-depth measurement technology.
Currently, being sounded the depth of the water by the method for remote sensing Depth extraction, the main method interpreted by theory, according to light at water
With the road radiation transmission process in air two media, set up from water radiation patterns based on the radiation field distribution in water body, and then
The quantitative model of information and water depth value is obtained to sensor.
Owing to the method for the remote sensing Depth extraction of theory interpretation model needs to calculate the depth of water according to all kinds of optical parametrics, and
All kinds of optical parametrics are difficult to directly measure, and numerical value has bigger error, the water depth value causing utilizing optical parametric to calculate also with
Having bigger error between the actual depth of water, and theoretical interpretation model is complex, method for solving is complicated and inefficient, is difficult to full
Foot engineering demand.
Summary of the invention
In view of this, the purpose of the embodiment of the present invention is to provide a kind of multispectral remote sensing depth of water based on nonlinear model
Inversion method and device, it is achieved based on actual measurement control point water depth value, by the non-linear inversion model solution to foundation, can be fast
Speed obtains the water depth value of the islands and reefs reef dish water field of big area away from continent, when setting up non-linear inversion model, only need to obtain optics
The reflectivity of the water surface of water field of big area in parameter, and the actual measurement control point water depth value of water field of big area be prone to measure, mould
Type is set up and solution procedure is simple, and quickly, and it is the highest to solve the water depth value precision obtained, the non-linear inversion of foundation for solving speed
Model is suitable for various Depth extraction engineering, and transplantability is fine.
First aspect, embodiments provides a kind of multispectral remote sensing Depth extraction side based on nonlinear model
Method, described method includes:
Obtain multi-spectrum remote sensing image and the actual measurement control point water depth value in default waters of predeterminable area, to described multispectral
Remote sensing image pre-processes, and obtains the reflectivity of predeterminable area;
According to threshold method based near infrared band Spectral Characteristic, described predeterminable area reflectivity is carried out land and water separation, obtain
Take the reflectivity of the water surface in default waters;
Reflectivity according to described water surface and described actual measurement control point water depth value, set up described default waters corresponding
Non-linear inversion model;
By the Stepwise Regression Algorithm, described non-linear inversion model is carried out recurrence process, described in after processing according to recurrence
Non-linear inversion model inversion obtains the water depth value in described default waters.
In conjunction with first aspect, embodiments provide the first possible implementation of above-mentioned first aspect, its
In, described described multi-spectrum remote sensing image is pre-processed, obtain predeterminable area reflectivity, including:
Described multi-spectrum remote sensing image is carried out geometric correction;
Multi-spectrum remote sensing image after described geometric correction is carried out radiation calibration and atmospheric correction, obtains described preset areas
Territory reflectivity.
In conjunction with first aspect, embodiments provide the implementation that the second of above-mentioned first aspect is possible, its
In, the described reflectivity according to described water surface and described actual measurement control point water depth value, set up described default waters corresponding
Non-linear inversion model, including:
Reflectivity according to described water surface and described actual measurement control point water depth value, that sets up shown in formula (1) is described
Preset the non-linear inversion model that waters is corresponding;
Wherein, in formula (1), Z is water depth value, m1For the constant for adjusting depth of water ratio, n is for being used for ensureing logarithm
Be on the occasion of and the constant of linear relationship, m0For the constant for compensating zero meter of depth of water, λi、λjIt is respectively wave band i and wave band j, RwFor
The reflectivity of water surface.
In conjunction with first aspect, embodiments provide the third possible implementation of above-mentioned first aspect, its
In, described by the Stepwise Regression Algorithm, described non-linear inversion model is carried out recurrence process, including:
The alternative factor of the Stepwise Regression Algorithm is chosen from non-linear inversion model;
Obtain standardization input data;
According to described standardization input data, significance test and Gauss Adam conversion, obtain the water depth effect factor;
Non-linear inversion model after non-linear inversion model containing the described water depth effect factor is processed as recurrence.
In conjunction with first aspect, embodiments provide the 4th kind of possible implementation of above-mentioned first aspect, its
In, after the described water depth value obtaining described default waters according to described non-linear inversion model, described method also includes:
Obtain the actual measurement water depth value at other control points presetting waters in addition to actual measurement control point;
Actual measurement water depth value according to other control points described and the water depth value in described default waters, calculate described default waters
Water depth value precision.
Second aspect, embodiments provides a kind of multispectral remote sensing Depth extraction based on nonlinear model dress
Putting, described device includes:
First acquisition module, for obtaining the multi-spectrum remote sensing image of predeterminable area and the actual measurement control point water in default waters
Deep value, pre-processes described multi-spectrum remote sensing image, obtains predeterminable area reflectivity;
Separation module, for entering described predeterminable area reflectivity according to threshold method based near infrared band Spectral Characteristic
Row land and water separates, and obtains the reflectivity of the water surface presetting waters;
Set up module, for the reflectivity according to described water surface and described actual measurement control point water depth value, set up described
Preset the non-linear inversion model that waters is corresponding;
Successive Regression module, for described non-linear inversion model being carried out recurrence process by the Stepwise Regression Algorithm,
Inverting module, the described non-linear inversion model inversion after processing according to recurrence obtains described default waters
Water depth value.
In conjunction with second aspect, embodiments provide the first possible implementation of above-mentioned second aspect, its
In, described first acquisition module includes:
First acquiring unit, the actual measurement of multi-spectrum remote sensing image and described default waters for obtaining predeterminable area controls
Point water depth value;
Pretreatment unit, for carrying out geometric correction to described multi-spectrum remote sensing image, to many after described geometric correction
Spectral remote sensing image carries out radiation calibration and atmospheric correction, obtains described predeterminable area reflectivity.
In conjunction with second aspect, embodiments provide the implementation that the second of above-mentioned second aspect is possible, its
In, described module of setting up includes:
Set up unit, for the reflectivity according to described water surface and described actual measurement control point water depth value, set up formula
(1) the non-linear inversion model that described default waters shown in is corresponding;
Wherein, in formula (1), Z is water depth value, m1For the constant for adjusting depth of water ratio, n is for being used for ensureing logarithm
Be on the occasion of and the constant of linear relationship, m0For the constant for compensating zero meter of depth of water, λi、λjIt is respectively wave band i and wave band j, RwFor
The reflectivity of water surface.
In conjunction with second aspect, embodiments provide the third possible implementation of above-mentioned second aspect, its
In, described successive Regression module includes:
Choose unit, for choosing the alternative factor of the Stepwise Regression Algorithm from non-linear inversion model;
Second acquisition unit, is used for obtaining standardization input data;
3rd acquiring unit, for according to described standardization input data, significance test and Gauss Adam conversion, obtaining
Fetch water deep factor of influence;
Determine unit, for using the non-linear inversion model containing the described water depth effect factor as recurrence process after non-
Linear inversion model.
In conjunction with second aspect, embodiments provide the 4th kind of possible implementation of above-mentioned second aspect, its
In, described device also includes:
Second acquisition module, for obtaining the actual measurement water depth value at other control points in addition to actual measurement control point, the default waters;
Computing module, for the actual measurement water depth value according to other control points described and the water depth value in described default waters, meter
Calculate the water depth value precision in described default waters.
In the method and device that the embodiment of the present invention provides, the method includes the multispectral remote sensing shadow obtaining predeterminable area
The actual measurement control point water depth value in picture and default waters, pre-processes multi-spectrum remote sensing image, obtains predeterminable area reflectivity;
According to threshold method based near infrared band Spectral Characteristic, predeterminable area reflectivity is carried out land and water separation, obtain and preset waters
The reflectivity of water surface;Reflectivity according to water surface and actual measurement control point water depth value, set up and preset corresponding non-in waters
Linear inversion model;By the Stepwise Regression Algorithm, non-linear inversion model is carried out recurrence process, according to recurrence process after non-
Linear inversion model inversion obtains presetting the water depth value in waters;This device includes the first acquisition module, is used for obtaining predeterminable area
Multi-spectrum remote sensing image and the actual measurement control point water depth value in default waters, multi-spectrum remote sensing image is pre-processed, obtains
Predeterminable area reflectivity;Separation module, for reflecting predeterminable area according to threshold method based near infrared band Spectral Characteristic
Rate carries out land and water separation, obtains the reflectivity of the water surface presetting waters;Set up module, for the reflection according to water surface
Rate and actual measurement control point water depth value, set up and preset the non-linear inversion model that waters is corresponding;Successive Regression module, for by by
Step regression algorithm carries out recurrence process to non-linear inversion model;Inverting module, for according to recurrence process after non-linear instead
Drill model inversion and obtain presetting the water depth value in waters.Realize based on actual measurement control point water depth value, by set up non-linear instead
Drill model solution, can quickly obtain the water depth value of the islands and reefs reef dish water field of big area away from continent, set up non-linear inversion mould
During type, only need to obtain the reflectivity of the water surface of water field of big area in optical parametric, and the actual measurement control point of water field of big area
Water depth value is prone to measure, and model is set up and solution procedure is simple, and solving speed quickly, and solves the water depth value precision obtained very
Height, the non-linear inversion model of foundation is suitable for various Depth extraction engineering, and transplantability is fine.
For making the above-mentioned purpose of the present invention, feature and advantage to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, below by embodiment required use attached
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, and it is right to be therefore not construed as
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to according to this
A little accompanying drawings obtain other relevant accompanying drawings.
Figure 1A shows a kind of based on nonlinear model the multispectral remote sensing Depth extraction that the embodiment of the present invention 1 is provided
The flow chart of method;
Figure 1B shows that the non-linear inversion model corresponding to default waters that the embodiment of the present invention 1 is provided returns
The flow chart processed;
Fig. 2 shows a kind of based on nonlinear model the multispectral remote sensing Depth extraction that the embodiment of the present invention 2 is provided
The structural representation of device.
Detailed description of the invention
Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground describes, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments.Generally exist
Can arrange and design with various different configurations with the assembly of the embodiment of the present invention that illustrates described in accompanying drawing herein.Cause
This, be not intended to limit claimed invention to the detailed description of the embodiments of the invention provided in the accompanying drawings below
Scope, but it is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, those skilled in the art are not doing
The every other embodiment obtained on the premise of going out creative work, broadly falls into the scope of protection of the invention.In view of existing
Carry out in the technology of Depth extraction based on theory interpretation model, need to calculate the depth of water according to all kinds of optical parametrics, and all kinds of light
Learning parameter to be difficult to directly measure, numerical value has bigger error, the water depth value causing utilizing optical parametric to calculate also with actual water
Having bigger error between Shen, and theoretical interpretation model is complex, method for solving is complicated and inefficient, is difficult to meet engineering
Demand.In consideration of it, the invention provides a kind of multispectral remote sensing inversion method based on nonlinear model and device, it is achieved
Based on actual measurement control point water depth value, by the non-linear inversion model solution set up, can quickly obtain the island away from continent
The water depth value of reef reef dish water field of big area, when setting up non-linear inversion model, only need to obtain water field of big area in optical parametric
The reflectivity of water surface, and the actual measurement control point water depth value of water field of big area be prone to measure, model set up and solution procedure letter
Single, quickly, and it is the highest to solve the water depth value precision obtained for solving speed, and it is anti-that the non-linear inversion model of foundation is suitable for the various depth of water
Drill engineering, and transplantability is fine.It is described below by embodiment.
Embodiment 1
See Figure 1A, embodiments provide a kind of multispectral remote sensing Depth extraction side based on nonlinear model
Method, the method includes following S101-S104 step.
Step S101: obtain multi-spectrum remote sensing image and the actual measurement control point water depth value in default waters of predeterminable area is right
Multi-spectrum remote sensing image pre-processes, and obtains predeterminable area reflectivity.
In embodiments of the present invention, above-mentioned default waters is the region, waters in above-mentioned predeterminable area, wherein, above-mentioned default
Region can be remote from the islands and reefs peripheral region in continent, and above-mentioned default waters can be remote from the water of the islands and reefs peripheral region in continent
Region, territory, in this islands and reefs reef dish actual measurement depth of water point, selected part is as the actual measurement control point of the depth of water.
Above-mentioned multi-spectrum remote sensing image can obtain by the way of satellite passes by, or obtains by the way of air remote sensing
Taking, wherein, the multi-spectrum remote sensing image resolution ratio got is the highest, ensure that based on high-resolution multi-spectrum remote sensing image
The water depth value precision that follow-up inverting obtains.
The actual measurement control point water depth value in above-mentioned default waters, can be obtained by the actual measurement of water-depth measurement equipment, thus obtain
The actual measurement control point water depth value accuracy in the default waters arrived is the highest.
After getting multi-spectrum remote sensing image, need multi-spectrum remote sensing image is pre-processed, from pretreated
Multi-spectrum remote sensing image obtains predeterminable area reflectivity.Wherein, pretreatment includes following S1011-S1012 step.
Step S1011: multi-spectrum remote sensing image is carried out geometric correction.
Step S1012: the multi-spectrum remote sensing image after geometric correction is carried out radiation calibration and atmospheric correction, is preset
Regional reflex rate.
In embodiments of the present invention, multi-spectrum remote sensing image is pre-processed, first multi-spectrum remote sensing image is carried out several
What correction, then carries out radiation calibration to multi-spectrum remote sensing image, finally multi-spectrum remote sensing image is carried out atmospheric correction, obtain
Predeterminable area reflectivity.
The depth of water obtained due to remote sensing Depth extraction is the sea absolute discrepancy in elevation to seabed, and surveying bathymetric data is with deeply
The depth of water on the basis of degree datum level is expressed, and the height datum of the two differs, and to be modeled and precision analysis, then needs reality
Survey bathymetric data and carry out elevation vertical datum transformation.So while above-mentioned multi-spectrum remote sensing image is pre-processed, also
Need actual measurement control point water depth value is carried out elevation vertical datum transformation and space interpolation, so that the water that remote sensing Depth extraction obtains
Deep value is identical with the height datum of actual measurement control point water depth value, finally obtains the most pre-according to actual measurement control point water depth value inverting
If the water depth value in waters.
After multi-spectrum remote sensing image is pre-processed, and actual measurement control point water depth value is carried out elevation vertical reference
After conversion and space interpolation, owing to predeterminable area includes waters and islands and reefs etc., predeterminable area reflectivity is more than presetting waters
The reflectivity of water surface, now can obtain the water body table presetting waters from predeterminable area reflectivity by following S102 step
The reflectivity in face.
Step S102: according to threshold method based near infrared band Spectral Characteristic, default waters reflectivity is carried out land and water and divide
From, obtain the reflectivity of the water surface presetting waters.
Threshold method, is based on image greyscale information according to setting threshold value and multi-spectrum remote sensing image is divided into different districts
Territory, if the image after Fen Ge be A (x, y), then segmentation after image be represented by A (x, y)=k, k=1,2,3 ... N, wherein,
In formula, K is cut zone code name.This threshold method is suitable for contrast more significantly image.Near-infrared shadow for land and water body
For Xiang, threshold method based near infrared band Spectral Characteristic is the Spectral Characteristic utilizing near infrared band, i.e. high anti-on land
Penetrating and the characteristic of water body almost hypersorption, separated in land and water body, wherein, it is right that above-mentioned multi-spectrum remote sensing image is converted into
The histogram answered shows, shows as the most bimodal, and reflectance curve valley occurs at flood boundaries.Wherein, by by this paddy
Value is set as segmentation threshold, is come in land and water segmentation according to this segmentation threshold, thus realizes land and water and separate, by islands and reefs and
Waters around islands and reefs separates, and gets the waters around islands and reefs, and then isolates islands and reefs from predeterminable area reflectivity
The reflectivity in waters around, i.e. presets the reflectivity of the water surface in waters.Should threshold based near infrared band Spectral Characteristic
Value method is simple, calculates speed quickly, and the reflectivity accuracy of the water surface in the default waters got is the highest.
After getting the reflectivity of water surface in default waters, can be set up by following S103 step and preset waters pair
The non-linear inversion model answered.
Step S103: according to the reflectivity of water surface and actual measurement control point water depth value, sets up and presets corresponding non-in waters
Linear inversion model.
By the reflectivity of water surface and actual measurement control point water depth value, presetting as shown in below equation (1) can be set up
The non-linear inversion model that waters is corresponding;
Wherein, in formula (1), Z is water depth value, m1For the constant for adjusting depth of water ratio, n is for being used for ensureing logarithm
Be on the occasion of and the constant of linear relationship, m0For the constant for compensating zero meter of depth of water, λi、λjIt is respectively wave band i and wave band j, RwFor
The reflectivity of water surface.
The non-linear inversion model that the default waters of above-mentioned foundation is corresponding, because of for same water body, although two wave bands are inhaled
Receipts coefficient is different, but decay difference is less, simultaneously can be by the reflectivity of the water surface of two wave bands is done ratio, two ripples
Section mutually compensates for because of the impact of different substrate reflectivity, the change of the different sediment types of the suppression impact on the depth of water, thus reduces
Non-linear inversion model error, obtains accurately calculating the non-linear inversion model of water depth value.
Owing to the non-linear inversion model that above-mentioned default waters is corresponding is the reflectivity according to water surface and actual measurement control
Point water depth value is set up, and wherein the actual measurement control point water depth value degree of accuracy is the highest, thus the non-linear inversion model accuracy set up
Height, by solving, can obtain the water depth value in the highest default waters of the degree of accuracy.
Owing to the non-linear inversion model that the default waters of above-mentioned foundation is corresponding is susceptible to harmful synteny problem, in order to
Quickly obtain wave band or the band combination that the depth of water is had a significant impact, improve the depth of water in the default waters that inverting obtains further
Value precision, it is also possible to by following S104 step, the non-linear inversion model that default waters is corresponding is carried out recurrence process, with excellent
Change the non-linear inversion model that this default waters is corresponding, thus obtain the higher water depth value of the degree of accuracy;Wherein, when certain wave band or
When band combination exceedes pre-set level to the impact of the depth of water, now illustrate that this wave band or band combination have conspicuousness shadow to the depth of water
Ringing, pre-set level can be the numerical value set.
Step S104: non-linear inversion model is carried out recurrence process by the Stepwise Regression Algorithm, after processing according to recurrence
Non-linear inversion model inversion obtain preset waters water depth value.
Wherein, by the Stepwise Regression Algorithm, non-linear inversion model is carried out recurrence process, following S1041-can be passed through
S1044 step processes, as shown in Figure 1B.
Step S1041: choose the alternative factor of the Stepwise Regression Algorithm from non-linear inversion model.
The above-mentioned alternative factor refers to the master variable in non-linear inversion model, and the foundation to non-linear inversion model has decision
Property impact factor.Such as, the reflectivity R of water surfacew。
Step S1042: obtain standardization input data.
When obtaining above-mentioned standardization input data, first according to normalized equation, input data are standardized and nothing
Dimension processes, and obtains standardization input data, wherein, inputs data, including the non-linear inversion mould that above-mentioned default waters is corresponding
The reflectivity of the water surface in type and the water depth value in calculated default waters, normalized equation is as shown in (2);
In equation (2), y is water depth value, and x is the reflectivity of water surface, and i is wave band, and t is the anti-of the water surface chosen
Penetrate the number of rate or water depth value,For the mean value of bathymetric data,For the mean value of the reflectivity of water surface, sZFor the depth of water
The standard deviation of value, siStandard deviation for water surface reflectivity;
After standardization,
In embodiments of the present invention, the purpose that input data are standardized, mainly eliminate the water surface of input
Reflectivity and water depth value, the problem being difficult to define regression coefficient dimension caused because of varying number level.
Step S1043: according to standardization input data, significance test and Gauss Adam conversion, obtain water depth effect
The factor.
The above-mentioned water depth effect factor, refers to the alternative factor having a significant impact the depth of water, wherein, when alternative factor pair water
When deep influence degree exceedes default value, now represent that this alternative factor pair depth of water has a significant impact;When alternative factor pair
When the influence degree of the depth of water is less than default value, now represent that this alternative factor pair depth of water does not has a significant impact.
Wherein, the water depth effect factor can be obtained by following S10431-S1043 step.
Step S10431: first input data Criterion equation group according to above-mentioned standardization, obtains being correlated with between variable
Coefficient matrix R (k), solves initial correlation matrix R (0) according to described R (k), R (0) as shown in (3),
Step S10432: calculate sum of squares of partial regression/or the contribution margin V being not introduced into the factor according to R (0)i (k), Vi (k)Table
Reach formula as shown in (4),
Above-mentioned it is not introduced into the factor, refers to the other factors in addition to the alternative factor.
Step S10433: when according to Vi (k)Expression formula determine maximum contribution value VmaxAfter the corresponding factor, by significantly
Property inspection formula F to VmaxThe corresponding factor carries out significance test, wherein, significance test formula F such as below equation (5) institute
Show,
If upchecking, performing following steps S10434, if the test fails, then performing following steps S10435.
Step S10434: set t as introducing or rejecting factor sequence number, then with rttCentered by, convert public affairs according to Gauss Adam
Formula carries out Gauss Adam's conversion to correlation matrix, wherein, shown in Gauss Adam's transformation for mula following (6);
In formula, k is kth time conversion, and i is the i-th row, and j is jth row;rijFor the i-th row in R (k), the element of jth row, t is for drawing
Entering or reject factor sequence number, often introduce a factor or reject a factor, described R (k) will convert;
Step S10435: repeated execution of steps S1041-S10434.When introduce wave band number more than or equal to 2 time, carry out by
Individual rejecting, obtains the factor minimum to depth of water contribution margin, if this factor is disallowable, performs step S10434.
When all Significance factors of the factor in non-linear inversion model, i.e. when not having the factor to introduce, there is no the factor yet
During rejecting, perform following S1044 step.
Step S1044: using the non-linear inversion model containing the water depth effect factor as recurrence process after non-linear instead
Drill model.
Above-mentioned the Stepwise Regression Algorithm is that one chooses optimum variable, solves the algorithm that harmful synteny problem is more ripe.
This algorithm includes introducing one by one and rejects two parts one by one, and general principle is first to introduce depth of water shadow with certain significance
Ringing the factor, such iterative cycles performs, and remains that being selected into the factor is the water depth effect factor.
Based on the Stepwise Regression Algorithm, non-linear inversion model is improved, both can ensure that the spy of master mould optimum variable
Property, make master mould the most quickly obtain wave band or the band combination that the depth of water is had a significant impact, further through introducing wave band to non-
M in linear inversion model0It is modified, replaces constant by band class information, improve variable in non-linear inversion model further
Precision, thus improve the water depth value precision in the default waters got.
After non-linear inversion model inversion after processing according to recurrence obtains the water depth value in default waters, it is also possible to meter
Calculate the water depth value precision in default waters.Including:
Obtain the actual measurement water depth value at other control points presetting waters in addition to actual measurement control point;
Actual measurement water depth value according to other control points and the water depth value in above-mentioned default waters, calculate the water depth value presetting waters
Precision.
In embodiments of the present invention, when the water depth value precision in waters is preset in above-mentioned calculating, RMSE (root-mean-can be passed through
Square error, root-mean-square error) formula calculates, wherein, shown in RMSE formula equation below (7).
In formula (7), ZiThe water depth value obtained for inverting,Surveying water depth value for correspondence position, n is selected by evaluation precision
Sample number.When calculating the water depth value precision presetting waters, by other in addition to actual measurement control point of the default waters that gets
The actual measurement water depth value at control pointAnd the water depth value Z in the default waters of correspondence positioniSubstitute into above-mentioned formula, such that it is able to
To water depth value precision RMSE presetting other control points, waters.
When, after the water depth value precision being calculated default waters, the water depth value precision in default waters being evaluated.
The RMSE value of above-mentioned calculating reflects the degree of the water depth value deviation actual measurement water depth value that inverting obtains, and wherein, RMSE value is the least, table
Showing that the degree of the water depth value deviation actual measurement water depth value obtained according to the non-linear inversion model inversion in the present invention is the least, inverting obtains
The water depth value precision arrived is the highest;RMSE value is the biggest, represents the depth of water obtained according to the non-linear inversion model inversion in the present invention
The degree of value deviation actual measurement water depth value is the biggest, and the water depth value precision that inverting obtains is the lowest.
When the water depth value precision that the non-linear inversion model inversion in the present invention obtains is the highest, then the present invention is passed through in explanation
The water depth value that the method provided calculates is very accurate, the suitable water depth value calculating the islands and reefs waters away from continent.
In the method that the embodiment of the present invention provides, a kind of multispectral remote sensing inversion method based on nonlinear model
Including multi-spectrum remote sensing image and the actual measurement control point water depth value in default waters of acquisition predeterminable area, to multi-spectrum remote sensing image
Pre-process, obtain predeterminable area reflectivity;Anti-to predeterminable area according to threshold method based near infrared band Spectral Characteristic
The rate of penetrating carries out land and water separation, obtains the reflectivity of the water surface presetting waters;Reflectivity according to water surface and actual measurement control
System point water depth value, sets up and presets the non-linear inversion model that waters is corresponding;By the Stepwise Regression Algorithm to non-linear inversion model
Carrying out recurrence process, the non-linear inversion model inversion after processing according to recurrence obtains presetting the water depth value in waters.Realize based on
Actual measurement control point water depth value, by the non-linear inversion model solution set up, can quickly obtain the islands and reefs reef away from continent
The water depth value of dish water field of big area, when setting up non-linear inversion model, only need to obtain the water body of water field of big area in optical parametric
The reflectivity on surface, and the actual measurement control point water depth value of water field of big area is prone to measure, model is set up and solution procedure is simple, asks
Solving speed quickly, and it is the highest to solve the water depth value precision obtained, the non-linear inversion model of foundation is suitable for various Depth extraction work
Journey, and transplantability is fine.
Embodiment 2
See Fig. 2, embodiments provide a kind of multispectral remote sensing Depth extraction device based on nonlinear model,
Device includes:
First acquisition module S1, for obtaining the multi-spectrum remote sensing image of predeterminable area and the actual measurement control point in default waters
Water depth value, pre-processes multi-spectrum remote sensing image, obtains predeterminable area reflectivity;
Separation module S2, carries out land and water for threshold method based near infrared band Spectral Characteristic to predeterminable area reflectivity
Separate, obtain the reflectivity of the water surface presetting waters;
Set up module S3, for the reflectivity according to water surface and actual measurement control point water depth value, set up and preset waters pair
The non-linear inversion model answered;
Successive Regression module S4, for carrying out recurrence process by the Stepwise Regression Algorithm to non-linear inversion model;
Inverting module S5, the non-linear inversion model inversion after processing according to recurrence obtains presetting the depth of water in waters
Value.
In the present embodiment, the actual measurement control point water depth value in above-mentioned multi-spectrum remote sensing image and default waters can be by above-mentioned
The method that embodiment 1 provides obtains, and does not repeats them here.
Above-mentioned first acquisition module S1 includes the first acquiring unit and pretreatment unit;
First acquiring unit, for obtaining the multi-spectrum remote sensing image of predeterminable area and the actual measurement control point water in default waters
Deep value;
Pretreatment unit, for carrying out geometric correction to multi-spectrum remote sensing image, to the multispectral remote sensing after geometric correction
Image carries out radiation calibration and atmospheric correction, obtains predeterminable area reflectivity.
Above-mentioned pretreatment unit first carries out geometric correction to multi-spectrum remote sensing image, then carries out radiation calibration, finally carries out
Data in multi-spectrum remote sensing image are carried out denoising and filtration by atmospheric correction, such that it is able to get predeterminable area accurately
Reflectivity;After actual measurement control point water depth value is carried out elevation vertical datum transformation and space interpolation, so that remote sensing Depth extraction obtains
The depth of water arrived is identical with the height datum of actual measurement control point water depth value, finally according to surveying the non-linear of control point water depth value foundation
Model is calculated inverting water depth value accurately.
After getting predeterminable area reflectivity, can be by the water surface in the default waters of above-mentioned separation module S2 acquisition
Reflectivity.Wherein, the reflectivity of the water surface that above-mentioned separation module S2 obtains default waters can be provided by above-described embodiment 1
Method obtain, do not repeat them here.
After the reflectivity of the water surface getting default waters as separation module S2, can be built by above-mentioned module S3 of setting up
The non-linear inversion model that vertical default waters is corresponding.
Above-mentioned module S3 of setting up includes setting up unit;
Set up unit, for the reflectivity according to water surface and actual measurement control point water depth value, set up shown in formula (1)
Preset the non-linear inversion model that waters is corresponding;
Wherein, in formula (1), Z is water depth value, m1For the constant for adjusting depth of water ratio, n is for being used for ensureing logarithm
Be on the occasion of and the constant of linear relationship, m0For the constant for compensating zero meter of depth of water, λi、λjIt is respectively wave band i and wave band j, RwFor
The reflectivity of water surface.
Above-mentioned unit of setting up processes reflectivity and the reality of the water surface obtained according to formula (1), above-mentioned pretreatment unit
Survey control point water depth value, set up and preset the non-linear inversion model that waters is corresponding.
The non-linear inversion model that the default waters of above-mentioned foundation is corresponding, because of for same water body, although two wave bands are inhaled
Receipts coefficient is different, but decay difference is less, simultaneously can be by the reflectivity of the water surface of two wave bands is done ratio, two ripples
Section mutually compensates for because of the impact of different substrate reflectivity, the change of the different sediment types of the suppression impact on the depth of water, thus reduces
Non-linear inversion model error, obtains accurately calculating the non-linear inversion model of water depth value.
Owing to the non-linear inversion model that above-mentioned default waters is corresponding is the reflectivity according to water surface and actual measurement control
Point water depth value is set up, and wherein the actual measurement control point water depth value degree of accuracy is the highest, thus the non-linear inversion model accuracy set up
Height, by solving, can obtain the water depth value in the highest default waters of the degree of accuracy.
Owing to the non-linear inversion model that the default waters of above-mentioned foundation is corresponding is susceptible to harmful synteny problem, in order to
Quickly obtain wave band or the band combination that the depth of water is had a significant impact, improve the depth of water in the default waters that inverting obtains further
Value precision, it is also possible to the non-linear inversion model that default waters is corresponding is carried out at recurrence by above-mentioned successive Regression module S4
Reason, the non-linear inversion model corresponding to optimize this default waters, thus obtain the higher water depth value of the degree of accuracy.
Above-mentioned successive Regression module S4 includes choosing unit, second acquisition unit, determining that unit, verification unit and the 3rd obtain
Take unit.
Choose unit, for choosing the alternative factor of the Stepwise Regression Algorithm from non-linear inversion model.
The above-mentioned alternative factor refers to the master variable in non-linear inversion model, and the foundation to non-linear inversion model has decision
Property impact factor.
Second acquisition unit, is used for obtaining standardization input data.
Above-mentioned second acquisition unit is when obtaining above-mentioned standardization input data, first according to normalized equation to input number
According to being standardized and dimensionless process, obtain standardization input data, wherein, input data, including above-mentioned default waters pair
The reflectivity of the water surface in the non-linear inversion model answered and the water depth value in calculated default waters, normalized equation
As shown in (2)
In equation (2), y is water depth value, and x is the reflectivity of water surface, and i is wave band, and t is the anti-of the water surface chosen
Penetrate the number of rate or water depth value,For the mean value of bathymetric data,For the mean value of the reflectivity of water surface, sZFor the depth of water
The standard deviation of value, siStandard deviation for water surface reflectivity;
After standardization,
In embodiments of the present invention, the purpose that input data are standardized, mainly eliminate the water surface of input
Reflectivity and water depth value, the problem being difficult to define regression coefficient dimension caused because of varying number level.
3rd acquiring unit, for according to standardization input data, significance test and Gauss Adam conversion, obtaining water
Deep factor of influence.
Wherein, the 3rd acquiring unit can obtain the water depth effect factor by the acquisition methods provided in above-described embodiment 1,
Do not repeat them here.
When all Significance factors of the factor in non-linear inversion model, i.e. when not having the factor to introduce, there is no the factor yet
During rejecting, obtain the non-linear inversion model after recurrence processes by unit identified below.
Determine unit, for using the non-linear inversion model containing the water depth effect factor as recurrence process after non-linear
Inverse model.
Non-linear inversion model is entered by the Stepwise Regression Algorithm that above-mentioned successive Regression module S4 provides according to above-described embodiment 1
Row recurrence processes.
Based on the Stepwise Regression Algorithm, non-linear inversion model is improved, both can ensure that the spy of master mould optimum variable
Property, make master mould the most quickly obtain wave band or the band combination that the depth of water is had a significant impact, further through introducing wave band to non-
M in linear inversion model0It is modified, replaces constant by band class information, improve variable in non-linear inversion model further
Precision, thus improve the water depth value precision in the default waters got.
After above-mentioned successive Regression module S4 carries out recurrence process to non-linear inversion model, above-mentioned inverting module S5 according to
Non-linear inversion model inversion after recurrence process obtains presetting the water depth value in waters.
After above-mentioned inverting module S5 inverting obtains the water depth value in default waters, also by following second acquisition module
The water depth value precision presetting waters is calculated with computing module, and according to the water depth value precision evaluation above-mentioned non-linear inversion mould calculated
Type.
Said apparatus also includes the second acquisition module and computing module;
Second acquisition module, for obtaining the actual measurement water depth value at other control points in addition to actual measurement control point, the default waters;
Computing module, for the actual measurement water depth value according to other control points and the water depth value in default waters, calculates and presets water
The water depth value precision in territory.
When above-mentioned computing module calculates the water depth value precision presetting waters, RMSE (root-mean-square can be passed through
Error, root-mean-square error) formula calculated, i.e. calculated by below equation (7).
In formula (7), ZiThe water depth value obtained for inverting,Surveying water depth value for correspondence position, n is selected by evaluation precision
Sample number.When calculating the water depth value precision presetting waters, by other in addition to actual measurement control point of the default waters that gets
The actual measurement water depth value at control pointAnd the water depth value Z in the default waters of correspondence positioniSubstitute into above-mentioned formula, such that it is able to
To water depth value precision RMSE presetting other control points, waters.
In embodiments of the present invention, when the water depth value precision in default waters is evaluated, according to above-mentioned computing module meter
The result calculated evaluates the water depth value precision in default waters.The RMSE value of above-mentioned calculating reflects the water depth value deviation that inverting obtains
The degree of actual measurement water depth value, wherein, RMSE value is the least, represents the water obtained according to the non-linear inversion model inversion in the present invention
The degree of deep value deviation actual measurement water depth value is the least, and the water depth value precision that inverting obtains is the highest;RMSE value is the biggest, represents according to this
The degree of the water depth value deviation actual measurement water depth value that the non-linear inversion model inversion in bright obtains is the biggest, the water depth value that inverting obtains
Precision is the lowest.
When the water depth value precision that the non-linear inversion model inversion in the present invention obtains is the highest, illustrate to be carried by the present invention
The water depth value that the method for confession calculates is very accurate, the suitable water depth value calculating the islands and reefs waters away from continent.
In the device that the embodiment of the present invention provides, a kind of multispectral remote sensing Depth extraction device based on nonlinear model
Including the first acquisition module, for obtaining the multi-spectrum remote sensing image of predeterminable area and the actual measurement control point depth of water in default waters
Value, pre-processes multi-spectrum remote sensing image, obtains predeterminable area reflectivity;Separation module, for based near infrared band
The threshold method of Spectral Characteristic carries out land and water separation to predeterminable area reflectivity, obtains the reflectivity of the water surface presetting waters;
Set up module, for according to the reflectivity of water surface and actual measurement control point water depth value, set up and preset corresponding non-linear in waters
Inverse model;Successive Regression module, for carrying out recurrence process by the Stepwise Regression Algorithm to non-linear inversion model;Inverting mould
Block, the non-linear inversion model inversion after processing according to recurrence obtains presetting the water depth value in waters.Realize based on actual measurement control
System point water depth value, by the non-linear inversion model solution set up, can quickly obtain big of the islands and reefs reef dish away from continent
The water depth value in ponding territory, when setting up non-linear inversion model, only need to obtain the water surface of water field of big area in optical parametric
Reflectivity, and the actual measurement control point water depth value of water field of big area is prone to measure, model is set up and solution procedure is simple, solving speed
Quickly, and it is the highest to solve the water depth value precision obtained, and the non-linear inversion model of foundation is suitable for various Depth extraction engineering, and moves
Planting property is fine.
The above, the only detailed description of the invention of the present invention, but protection scope of the present invention is not limited thereto, and any
Those familiar with the art, in the technical scope that the invention discloses, can readily occur in change or replace, should contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention should described be as the criterion with scope of the claims.
Claims (10)
1. a multispectral remote sensing inversion method based on nonlinear model, it is characterised in that described method includes:
Obtain multi-spectrum remote sensing image and the actual measurement control point water depth value in default waters of predeterminable area, to described multispectral remote sensing
Image pre-processes, and obtains predeterminable area reflectivity;
According to threshold method based near infrared band Spectral Characteristic, described predeterminable area reflectivity is carried out land and water separation, obtain pre-
If the reflectivity of the water surface in waters;
Reflectivity according to described water surface and described actual measurement control point water depth value, set up the non-thread that described default waters is corresponding
Property inverse model;
By the Stepwise Regression Algorithm, described non-linear inversion model is carried out recurrence process, the described non-thread after processing according to recurrence
Property inverse model inverting obtains the water depth value in described default waters.
Method the most according to claim 1, it is characterised in that described described multi-spectrum remote sensing image is pre-processed,
Obtain predeterminable area reflectivity, including:
Described multi-spectrum remote sensing image is carried out geometric correction;
Multi-spectrum remote sensing image after described geometric correction is carried out radiation calibration and atmospheric correction, obtains described predeterminable area anti-
Penetrate rate.
Method the most according to claim 1, it is characterised in that the described reflectivity according to described water surface and described reality
Survey control point water depth value, set up the non-linear inversion model that described default waters is corresponding, including:
Reflectivity according to described water surface and described actual measurement control point water depth value, set up shown in formula (1) described preset
The non-linear inversion model that waters is corresponding;
Wherein, in formula (1), Z is water depth value, m1For the constant for adjusting depth of water ratio, n is for being used for ensureing that logarithm is just
Value and the constant of linear relationship, m0For the constant for compensating zero meter of depth of water, λi、λjIt is respectively wave band i and wave band j, RwFor water body
The reflectivity on surface.
Method the most according to claim 1, it is characterised in that described by the Stepwise Regression Algorithm to described non-linear inversion
Model carries out recurrence process, including:
The alternative factor of the Stepwise Regression Algorithm is chosen from non-linear inversion model;
Obtain standardization input data;
According to described standardization input data, significance test and Gauss Adam conversion, obtain the water depth effect factor;
Non-linear inversion model after non-linear inversion model containing the described water depth effect factor is processed as recurrence.
Method the most according to claim 1, it is characterised in that described described pre-according to the acquisition of described non-linear inversion model
If after the water depth value in waters, described method also includes:
Obtain the actual measurement water depth value at other control points presetting waters in addition to actual measurement control point;
Actual measurement water depth value according to other control points described and the water depth value in described default waters, calculate the water in described default waters
Deeply it is worth precision.
6. a multispectral remote sensing Depth extraction device based on nonlinear model, it is characterised in that described device includes:
First acquisition module, for obtaining the multi-spectrum remote sensing image of predeterminable area and the actual measurement control point depth of water in default waters
Value, pre-processes described multi-spectrum remote sensing image, obtains predeterminable area reflectivity;
Separation module, for carrying out water according to threshold method based near infrared band Spectral Characteristic to described predeterminable area reflectivity
Land separates, and obtains the reflectivity of the water surface presetting waters;
Set up module, for the reflectivity according to described water surface and described actual measurement control point water depth value, set up described presetting
The non-linear inversion model that waters is corresponding;
Successive Regression module, for described non-linear inversion model being carried out recurrence process by the Stepwise Regression Algorithm,
Inverting module, the described non-linear inversion model inversion after processing according to recurrence obtains the depth of water in described default waters
Value.
Device the most according to claim 6, it is characterised in that described first acquisition module includes:
First acquiring unit, the actual measurement of multi-spectrum remote sensing image and described default waters for obtaining described predeterminable area controls
Point water depth value;
Pretreatment unit, for described multi-spectrum remote sensing image is carried out geometric correction, multispectral to after described geometric correction
Remote sensing image carries out radiation calibration and atmospheric correction, obtains described predeterminable area reflectivity.
Device the most according to claim 6, it is characterised in that described module of setting up includes:
Set up unit, for the reflectivity according to described water surface and described actual measurement control point water depth value, set up formula (1) institute
The non-linear inversion model that the described default waters that shows is corresponding;
Wherein, in formula (1), Z is water depth value, m1For for adjusting the constant of depth of water ratio, n for be used for ensureing logarithm be on the occasion of
And the constant of linear relationship, m0For the constant for compensating zero meter of depth of water, λi、λjIt is respectively wave band i and wave band j, RwFor water body table
The reflectivity in face.
Device the most according to claim 6, it is characterised in that described successive Regression module includes:
Choose unit, for choosing the alternative factor of the Stepwise Regression Algorithm from non-linear inversion model;
Second acquisition unit, is used for obtaining standardization input data;
3rd acquiring unit, for according to described standardization input data, significance test and Gauss Adam conversion, obtaining water
Deep factor of influence;
Determine unit, for using the non-linear inversion model containing the described water depth effect factor as recurrence process after non-linear
Inverse model.
Device the most according to claim 6, it is characterised in that described device also includes:
Second acquisition module, for obtaining the actual measurement water depth value at other control points in addition to actual measurement control point, the default waters;
Computing module, for the actual measurement water depth value according to other control points described and the water depth value in described default waters, calculates institute
State the water depth value precision in default waters.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610228558.4A CN105865424B (en) | 2016-04-13 | 2016-04-13 | A kind of multispectral remote sensing inversion method and device based on nonlinear model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610228558.4A CN105865424B (en) | 2016-04-13 | 2016-04-13 | A kind of multispectral remote sensing inversion method and device based on nonlinear model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105865424A true CN105865424A (en) | 2016-08-17 |
CN105865424B CN105865424B (en) | 2019-09-17 |
Family
ID=56636927
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610228558.4A Active CN105865424B (en) | 2016-04-13 | 2016-04-13 | A kind of multispectral remote sensing inversion method and device based on nonlinear model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105865424B (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106950197A (en) * | 2017-03-03 | 2017-07-14 | 环境保护部卫星环境应用中心 | The Remotely sensed acquisition methods, devices and systems of sewage draining exit polluted-water |
CN108286962A (en) * | 2018-01-31 | 2018-07-17 | 中国科学院遥感与数字地球研究所 | A kind of method for building up and system of water environment library of spectra |
CN109059796A (en) * | 2018-07-20 | 2018-12-21 | 国家海洋局第三海洋研究所 | The multispectral satellite remote sensing inversion method of shallow water depth without depth of water control point region |
CN110132246A (en) * | 2019-04-03 | 2019-08-16 | 自然资源部第一海洋研究所 | RS Fathoming detection method based on residual error subregion |
CN110823190A (en) * | 2019-09-30 | 2020-02-21 | 广州地理研究所 | Island reef shallow sea water depth prediction method based on random forest |
CN110849334A (en) * | 2019-09-30 | 2020-02-28 | 广州地理研究所 | Island reef shallow sea water depth prediction method based on classification and regression tree |
CN110849333A (en) * | 2019-09-30 | 2020-02-28 | 广州地理研究所 | Island shallow sea water depth prediction method based on neural network |
CN110851789A (en) * | 2019-09-30 | 2020-02-28 | 广州地理研究所 | Island reef shallow sea water depth prediction method based on extreme gradient lifting |
CN111561916A (en) * | 2020-01-19 | 2020-08-21 | 自然资源部第二海洋研究所 | Shallow sea water depth uncontrolled extraction method based on four-waveband multispectral remote sensing image |
CN111829957A (en) * | 2020-07-07 | 2020-10-27 | 塔里木大学 | System and method for inverting moisture content of winter wheat plants based on multispectral remote sensing of unmanned aerial vehicle |
CN111879709A (en) * | 2020-07-15 | 2020-11-03 | 中国科学院空天信息创新研究院 | Method and device for detecting spectral reflectivity of lake water body |
CN111947628A (en) * | 2020-08-25 | 2020-11-17 | 自然资源部第一海洋研究所 | Linear water depth inversion method based on inherent optical parameters |
CN112013822A (en) * | 2020-07-22 | 2020-12-01 | 武汉智图云起科技有限公司 | Multispectral remote sensing water depth inversion method based on improved GWR model |
CN112489108A (en) * | 2020-12-24 | 2021-03-12 | 中国科学院南海海洋研究所 | Method and device for inverting and reconstructing underwater surface area of open-sea coral reef |
CN114459438A (en) * | 2022-01-10 | 2022-05-10 | 山东科技大学 | Method for judging validity of high-resolution multispectral water depth inversion data based on spectral roughness information |
CN114758254A (en) * | 2022-06-15 | 2022-07-15 | 中国地质大学(武汉) | Dual-band unsupervised water depth inversion method and system |
CN117372891A (en) * | 2023-12-07 | 2024-01-09 | 中铁水利水电规划设计集团有限公司 | Method for carrying out water depth inversion by using satellite remote sensing image |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SU956983A1 (en) * | 1980-04-17 | 1982-09-07 | Коммунарский горно-металлургический институт | Water reservoir depth determination method |
CN102589528A (en) * | 2012-01-19 | 2012-07-18 | 中国测绘科学研究院 | Multi-temporal imaging island shoreline surveying method |
CN103293521A (en) * | 2013-06-26 | 2013-09-11 | 武汉大学 | Method for detecting water depth of offshore sea by X-band radar |
CN104613944A (en) * | 2015-01-27 | 2015-05-13 | 电子科技大学 | Distributed water depth prediction method based on GWR (geographically weighted regression) and BP (back propagation) neural network |
CN104749579A (en) * | 2015-04-02 | 2015-07-01 | 太原理工大学 | Channel depth measuring method based on chaotic laser device and correlative method thereof |
CN105445751A (en) * | 2015-11-18 | 2016-03-30 | 国家海洋局第一海洋研究所 | Shallow water area water depth ratio remote sensing inversion method |
-
2016
- 2016-04-13 CN CN201610228558.4A patent/CN105865424B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SU956983A1 (en) * | 1980-04-17 | 1982-09-07 | Коммунарский горно-металлургический институт | Water reservoir depth determination method |
CN102589528A (en) * | 2012-01-19 | 2012-07-18 | 中国测绘科学研究院 | Multi-temporal imaging island shoreline surveying method |
CN103293521A (en) * | 2013-06-26 | 2013-09-11 | 武汉大学 | Method for detecting water depth of offshore sea by X-band radar |
CN104613944A (en) * | 2015-01-27 | 2015-05-13 | 电子科技大学 | Distributed water depth prediction method based on GWR (geographically weighted regression) and BP (back propagation) neural network |
CN104749579A (en) * | 2015-04-02 | 2015-07-01 | 太原理工大学 | Channel depth measuring method based on chaotic laser device and correlative method thereof |
CN105445751A (en) * | 2015-11-18 | 2016-03-30 | 国家海洋局第一海洋研究所 | Shallow water area water depth ratio remote sensing inversion method |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106950197A (en) * | 2017-03-03 | 2017-07-14 | 环境保护部卫星环境应用中心 | The Remotely sensed acquisition methods, devices and systems of sewage draining exit polluted-water |
CN108286962B (en) * | 2018-01-31 | 2019-11-05 | 中国科学院遥感与数字地球研究所 | A kind of method for building up and system of water environment library of spectra |
CN108286962A (en) * | 2018-01-31 | 2018-07-17 | 中国科学院遥感与数字地球研究所 | A kind of method for building up and system of water environment library of spectra |
CN109059796A (en) * | 2018-07-20 | 2018-12-21 | 国家海洋局第三海洋研究所 | The multispectral satellite remote sensing inversion method of shallow water depth without depth of water control point region |
CN110132246A (en) * | 2019-04-03 | 2019-08-16 | 自然资源部第一海洋研究所 | RS Fathoming detection method based on residual error subregion |
CN110132246B (en) * | 2019-04-03 | 2021-07-02 | 自然资源部第一海洋研究所 | Water depth remote sensing detection method based on residual partitioning |
CN110823190A (en) * | 2019-09-30 | 2020-02-21 | 广州地理研究所 | Island reef shallow sea water depth prediction method based on random forest |
CN110849334A (en) * | 2019-09-30 | 2020-02-28 | 广州地理研究所 | Island reef shallow sea water depth prediction method based on classification and regression tree |
CN110851789A (en) * | 2019-09-30 | 2020-02-28 | 广州地理研究所 | Island reef shallow sea water depth prediction method based on extreme gradient lifting |
CN110849333A (en) * | 2019-09-30 | 2020-02-28 | 广州地理研究所 | Island shallow sea water depth prediction method based on neural network |
CN111561916A (en) * | 2020-01-19 | 2020-08-21 | 自然资源部第二海洋研究所 | Shallow sea water depth uncontrolled extraction method based on four-waveband multispectral remote sensing image |
CN111561916B (en) * | 2020-01-19 | 2021-09-28 | 自然资源部第二海洋研究所 | Shallow sea water depth uncontrolled extraction method based on four-waveband multispectral remote sensing image |
CN111829957A (en) * | 2020-07-07 | 2020-10-27 | 塔里木大学 | System and method for inverting moisture content of winter wheat plants based on multispectral remote sensing of unmanned aerial vehicle |
CN111879709A (en) * | 2020-07-15 | 2020-11-03 | 中国科学院空天信息创新研究院 | Method and device for detecting spectral reflectivity of lake water body |
CN111879709B (en) * | 2020-07-15 | 2023-05-30 | 中国科学院空天信息创新研究院 | Lake water body spectral reflectivity inspection method and device |
CN112013822A (en) * | 2020-07-22 | 2020-12-01 | 武汉智图云起科技有限公司 | Multispectral remote sensing water depth inversion method based on improved GWR model |
CN111947628A (en) * | 2020-08-25 | 2020-11-17 | 自然资源部第一海洋研究所 | Linear water depth inversion method based on inherent optical parameters |
CN112489108A (en) * | 2020-12-24 | 2021-03-12 | 中国科学院南海海洋研究所 | Method and device for inverting and reconstructing underwater surface area of open-sea coral reef |
CN112489108B (en) * | 2020-12-24 | 2023-08-01 | 中国科学院南海海洋研究所 | Method and device for reconstructing underwater surface area inversion of open sea coral reef |
CN114459438A (en) * | 2022-01-10 | 2022-05-10 | 山东科技大学 | Method for judging validity of high-resolution multispectral water depth inversion data based on spectral roughness information |
CN114459438B (en) * | 2022-01-10 | 2024-02-02 | 山东科技大学 | Method for judging validity of high-resolution multispectral water depth inversion data |
CN114758254A (en) * | 2022-06-15 | 2022-07-15 | 中国地质大学(武汉) | Dual-band unsupervised water depth inversion method and system |
CN117372891A (en) * | 2023-12-07 | 2024-01-09 | 中铁水利水电规划设计集团有限公司 | Method for carrying out water depth inversion by using satellite remote sensing image |
CN117372891B (en) * | 2023-12-07 | 2024-02-13 | 中铁水利水电规划设计集团有限公司 | Method for carrying out water depth inversion by using satellite remote sensing image |
Also Published As
Publication number | Publication date |
---|---|
CN105865424B (en) | 2019-09-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105865424A (en) | Nonlinear model-based multispectral remote sensing water depth inversion method and apparatus thereof | |
CN104613944B (en) | Distributed water depth prediction method based on GWR (geographically weighted regression) and BP (back propagation) neural network | |
CN111476159B (en) | Method and device for training and detecting detection model based on double-angle regression | |
CN105445751B (en) | A kind of shallow water area depth of water ratio remote sensing inversion method | |
CN112712535B (en) | Mask-RCNN landslide segmentation method based on simulation difficult sample | |
CN109146889A (en) | A kind of field boundary extracting method based on high-resolution remote sensing image | |
CN114139444B (en) | Offshore sea surface temperature inversion method based on machine learning | |
CN104318051B (en) | The rule-based remote sensing of Water-Body Information on a large scale automatic extracting system and method | |
CN102063720B (en) | Treelets-based method for detecting remote sensing image changes | |
CN106600595A (en) | Human body characteristic dimension automatic measuring method based on artificial intelligence algorithm | |
Rahimi et al. | Locating sinkholes in LiDAR coverage of a glacio-fluvial karst, Winona County, MN | |
CN104156629B (en) | A kind of pathfinder image inverting wind direction of ocean surface method based on relative detector calibration | |
CN104050643B (en) | Remotely sensing image geometric and radiation integrated relative correction method and system | |
CN111598098A (en) | Water gauge water line detection and effectiveness identification method based on full convolution neural network | |
Osorio et al. | An algorithm for the measurement of shoreline and intertidal beach profiles using video imagery: PSDM | |
CN104732520A (en) | Cardio-thoracic ratio measuring algorithm and system for chest digital image | |
CN106920235A (en) | Star-loaded optical remote sensing image automatic correction method based on the matching of vector base map | |
CN109557030A (en) | A kind of water quality element inversion method based on unmanned aerial vehicle remote sensing | |
CN110163287A (en) | A kind of mesoscale eddy detection method and device | |
CN105241428A (en) | Water depth retrieval method with hyper-spectrum | |
CN114297938A (en) | Optical shallow water bottom depth inversion method based on neural network | |
CN104809471A (en) | Hyperspectral image residual error fusion classification method based on space spectrum information | |
CN106940782A (en) | High score SAR based on variogram increases construction land newly and extracts software | |
Chen et al. | An intelligent vision recognition method based on deep learning for pointer meters | |
CN110927065A (en) | Remote sensing assisted lake and reservoir chl-a concentration spatial interpolation method optimization method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |