CN107609578A - Remote sensing image texture analysis method based on multi-scale wavelet decomposition and fractal theory - Google Patents
Remote sensing image texture analysis method based on multi-scale wavelet decomposition and fractal theory Download PDFInfo
- Publication number
- CN107609578A CN107609578A CN201710738940.4A CN201710738940A CN107609578A CN 107609578 A CN107609578 A CN 107609578A CN 201710738940 A CN201710738940 A CN 201710738940A CN 107609578 A CN107609578 A CN 107609578A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- sensing image
- wavelet decomposition
- sample
- texture analysis
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention belongs to remote sensing image processing and applied technical field, a kind of Remote Sensing Image Texture analysis method based on multi-scale wavelet decomposition and fractal theory is specifically disclosed, a kind of Remote Sensing Image Texture analysis method based on multi-scale wavelet decomposition and fractal theory, this method comprise the following steps：Step 1, remote sensing image data is obtained；Step 2, the different lithology sample image on remote sensing image obtained to step 1 is chosen；Step 3, gray processing is carried out to the sample image that step 2 is chosen, obtains sample gray level image；Step 4, the different lithology sample gray level image obtained to step 3 carries out biorthogonal wavelet decomposition；Step 5, different stage small echo high frequency, the box meter dimension values and multifractal spectra of low frequency signal after being decomposed respectively in calculation procedure 4, complete Remote Sensing Image Texture analysis.This method utilizes different stage small echo high frequency, the box meter dimension and multifractal spectra parameter of low frequency signal after decomposing, quantitative analysis remote sensing images particular texture.
Description
Technical field
The invention belongs to remote sensing image processing and applied technical field, and in particular to one kind based on multi-scale wavelet decomposition and
The Remote Sensing Image Texture analysis method of fractal theory.
Background technology
Texture is one of key character of remote sensing images, it disclose in image the structural information of atural object and its with ring around
The relation in border, there is provided the important information of ground mulching type space change.The texture of remote sensing images is mainly shown as atural object
Spatial relationship and luminance contrast relation between shape, size, orientation, homogeneous degree and different atural objects etc..In geology field
In, different rocks has different surface characteristics because its distinctive mineral composition and structure construct, and these features exist
Different tones, shadow line pattern and drainage characteristic are shown as on remote sensing image.By long-term image interpretation and discovery is summarized, no
With the natural surface that is formed of rock type have nothing in common with each other, as sedimentary rock shows shallower in the picture, lineation is obvious；Volcano
The texture light and shade of rock is clearly demarcated, and directional spreding is uniform；The grain distribution of granite has obvious directionality.Therefore, by visual solution
Lithology can be distinguished to a certain extent by translating, but certain subjectivity be present, and remote sensing geology educational circles wishes to find one kind more always
Add intelligent texture analysis method, the texture and structural characteristic of quantitative judge different type cartographic feature.
Fractal method has ripe Fundamentals of Mathematics, it can deeply, reflect natural structure and geometric properties exactly, and
All multi-parameters in terms of geometric properties are characterized are made every effort to disclose its Nonlinear Dynamics, are that research remote sensing image texture is special
The effective tool of sign.It has been widely used in the simulation of the true landscape of the earth's surfaces such as mountains and rivers, river, cloud at present.
Remote Sensing Image Texture visual interpretation disturbing factor is more at present and certain subjectivity be present, lacks Quantitative Analysis Method
The problems such as, need a kind of analysis method for quantitatively portraying different type rock image texture characteristic of research badly.
The content of the invention
The present invention solves the problems, such as：Overcome remote sensing images visual interpretation and traditional Remote Sensing Image Texture analysis method
Deficiency, there is provided a kind of Remote Sensing Image Texture analysis method based on multi-scale wavelet decomposition and fractal theory, after decomposition
The box meter dimension and multifractal spectra parameter of different stage small echo high frequency, low frequency signal, quantitatively portray Remote Sensing Image Texture, so as to
Quantitative analysis remote sensing images particular texture.
Realize the technical scheme of the object of the invention：A kind of remote sensing images line based on multi-scale wavelet decomposition and fractal theory
Analysis method is managed, this method comprises the following steps：
Step 1, remote sensing image data is obtained；
Step 2, the different lithology sample image on remote sensing image obtained to step 1 is chosen；
Step 3, gray processing is carried out to the sample image that step 2 is chosen, obtains sample gray level image；
Step 4, the different lithology sample gray level image obtained to step 3 carries out biorthogonal wavelet decomposition；
Step 5, the box meter dimension values of different stage small echo high frequency, low frequency signal after decomposing respectively in calculation procedure 4 and
Multifractal spectra, complete Remote Sensing Image Texture analysis.
The cloud coverage of remote sensing image data in described step 1 is less than 10%.
A kind of Remote Sensing Image Texture analysis method based on multi-scale wavelet decomposition and fractal theory, it is characterised in that：Institute
Remote sensing image data in the step 1 stated adds up to 7 multi light spectrum hands, and the spatial resolution of 1-5 wave bands and 7 wave bands is 30m,
Select 7,4,1 wave band to carry out RGB color synthesis, obtain chromatic image.
3 metamorphic rocks, volcanic rock and granite are chosen respectively on the remote sensing image that step 1 obtains in described step 2
Sample image, it is desirable to chosen under the conditions of equal proportion chi, the lithology classification in same sample image is consistent and selected
All sample image sizes it is identical.
With 1 in described step 2:200000 geologic maps carry out sample image selection as reference.
The different lithology sample image of selection is carried out by the function rgb2gray of matlab softwares in described step 3
Gray processing processing, obtains lithology sample gray level image.
The specific formula of lithology sample gray level image is as follows in described step 3：
Gray=0.2989*R+0.5870*G+0.1140*B.
Select the 2D signal progress of wavelet2D function pair sample images in matlab softwares more in described step 4
Yardstick discrete wavelet transformation, Selection of Wavelet Basis bior2.2 biorthogonal wavelets, multi-scale wavelet decomposition to the third level.
Pass through the 1st, 2,3 grade of small wave height after wavelet decomposition in FRACLAB softwares difference calculation procedure 4 in described step 5
Frequently, the box meter dimension values and multifractal spectra of low frequency signal.
Precision parameter is set as 0.40 during being calculated in described step 5, and dimension computational methods choose box meter dimension, area
Domain gray-scale statistical method chooses foundation that is using selection area gray value and being calculated as dimension.
The advantageous effects of the present invention are：The present invention proposes one kind and combines multi-scale wavelet decomposition and fractal theory
Remote Sensing Image Texture analysis method, by the remote sensing image of acquisition, different lithology sample image is chosen according to geologic map information,
And by sample image gray processing, obtain gray level image；Then biorthogonal wavelet decomposition is carried out to different lithology sample gray level image,
And to the small echo high frequency of different stage, low frequency signal calculation box meter dimension values and multifractal spectra parameter after decomposition, quantitatively portray
The textural characteristics of remote sensing images, overcome remote sensing images visual interpretation method disturbing factor more and certain subjectivity be present, and
Traditional texture analysis method carries out statistical analysis just for original image, portrays the problems such as not comprehensive enough.It is multiple dimensioned by introducing
Wavelet decomposition and fractal method, it can more fully reflect Remote Sensing Image Texture architectural characteristic, the texture obtained under multiple dimensioned
Feature difference, it will classify for Remote Sensing Image Texture and Lithology Discrimination provides important references.
Brief description of the drawings
Fig. 1 is provided by the present invention a kind of based on the analysis of the Remote Sensing Image Texture of multi-scale wavelet decomposition and fractal theory
The flow chart of method.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.
As shown in figure 1, the present invention be it is a kind of based on the Remote Sensing Image Texture of multi-scale wavelet decomposition and fractal theory analyze
Method, this method comprise the following steps：
Step 1, remote sensing image data is obtained
Obtain the remote sensing image data of desired zone, it is desirable to which cloud coverage is less than 10%.The ETM data that the present invention uses are total
In respect of 7 multi light spectrum hands, the spatial resolution of 1-5 wave bands and 7 wave bands is 30m, selects 7,4,1 wave band to carry out RGB color conjunction
Into obtaining chromatic image.
ETM (Enhanced Thematic Mapper), Enhanced Thematic Mapper；RGB represents red (R), green (G), indigo plant
(B) color of three passages.
Step 2, the different lithology sample image on remote sensing image obtained to step 1 is chosen, and specifically includes：
With 1:200000 geologic maps choose 3 metamorphic rocks, volcanoes respectively as reference on the remote sensing image that step 1 obtains
Rock and granite sample image, it is desirable to chosen under the conditions of equal proportion chi, the lithology classification in same sample image needs
It is consistent, and selected all sample image sizes are identical, each sample image ranks number is close.
Step 3, gray processing is carried out to the sample image that step 2 is chosen, obtains sample gray level image, specifically include：
The different lithology sample image of selection is carried out at gray processing by the function rgb2gray that matlab softwares carry
Reason, obtains lithology sample gray level image, the specific formula of lithology sample gray level image is as follows：
Gray=0.2989*R+0.5870*G+0.1140*B
Wherein, gray represents grayscale image values, and R, G, B represent three wave bands of original sample image red, green, blue respectively
DN values.
Step 4, the different lithology sample gray level image obtained to step 3 carries out biorthogonal wavelet decomposition, specifically includes：
The 2D signal for the sample image that wavelet2D function pair steps 3 obtain carries out more chis in selection matlab softwares
Spend discrete wavelet transformation, Selection of Wavelet Basis bior2.2 biorthogonal wavelets, because biorthogonal wavelet has linear phase, extensively
It is preferable applied to signal and Image Reconstruction, effect.With the increase of wavelet decomposition series, the information included per one-level substantially subtracts
Few, the otherness of reflection different lithology texture structure is also reducing, therefore only accomplishes the third level for multi-scale wavelet decomposition.
Step 5, the box meter dimension values of different stage small echo high frequency, low frequency signal after decomposing respectively in calculation procedure 4 and
Multifractal spectra, comprise the following steps that：
1st, 2,3 grade of small echo high frequency after wavelet decomposition, low frequency signal are distinguished in calculation procedure 4 by FRACLAB softwares
Box meter dimension values and multifractal spectra, precision parameter is set as 0.40 in calculating process, and dimension computational methods choose box meter dimension,
Area grayscale statistical method chooses sum, i.e., chooses foundation that is selection area gray value and being calculated as dimension, other specification
Software default value.
Found according to result of calculation, the box meter dimension values of three kinds of lithology sample images have differences, on the whole metamorphic rock box
Dimension values highest is counted, volcanic rock takes second place, and granite is minimum, and same lithology sample low frequency signal box meter dimension values are more than high frequency
The box meter dimension values of signal.With the increase of decomposed class, the high-frequency signal and corresponding low frequency letter of three kinds of lithology sample images
Number box meter dimension values are presented the trend reduced, and their multifractal spectra in some sections there is also significant difference, this
The textural characteristics difference obtained a bit in different scale, important references are provided for Remote Sensing Image Texture classification and Lithology Discrimination.
The present invention is explained in detail above in conjunction with drawings and examples, but the present invention is not limited to above-mentioned implementation
Example, in those of ordinary skill in the art's possessed knowledge, can also make on the premise of present inventive concept is not departed from
Go out various change.The content not being described in detail in the present invention can use prior art.
Claims (10)
- A kind of 1. Remote Sensing Image Texture analysis method based on multi-scale wavelet decomposition and fractal theory, it is characterised in that the party Method comprises the following steps：Step 1, remote sensing image data is obtained；Step 2, the different lithology sample image on remote sensing image obtained to step 1 is chosen；Step 3, gray processing is carried out to the sample image that step 2 is chosen, obtains sample gray level image；Step 4, the different lithology sample gray level image obtained to step 3 carries out biorthogonal wavelet decomposition；Step 5, box meter dimension values of different stage small echo high frequency, low frequency signal after decomposing respectively in calculation procedure 4 and multiple Divide shape spectrum, complete Remote Sensing Image Texture analysis.
- A kind of 2. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 1 Method, it is characterised in that：The cloud coverage of remote sensing image data in described step 1 is less than 10%.
- A kind of 3. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 2 Method, it is characterised in that：Remote sensing image data in described step 1 adds up to 7 multi light spectrum hands, 1-5 wave bands and 7 wave bands Spatial resolution is 30m, selects 7,4,1 wave band to carry out RGB color synthesis, obtains chromatic image.
- A kind of 4. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 3 Method, it is characterised in that：In described step 2 step 1 obtain remote sensing image on choose respectively 3 metamorphic rocks, volcanic rock and Granite sample image, it is desirable to be chosen under the conditions of equal proportion chi, the lithology classification in same sample image is consistent, and Selected all sample image sizes are identical.
- A kind of 5. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 4 Method, it is characterised in that：With 1 in described step 2:200000 geologic maps carry out sample image selection as reference.
- A kind of 6. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 5 Method, it is characterised in that：Different lithology sample graph in described step 3 by the function rgb2gray of matlab softwares to selection As carrying out gray processing processing, lithology sample gray level image is obtained.
- A kind of 7. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 6 Method, it is characterised in that：The specific formula of lithology sample gray level image is as follows in described step 3：Gray=0.2989*R+0.5870*G+0.1140*B.
- A kind of 8. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 7 Method, it is characterised in that：The 2D signal of wavelet2D function pair sample images in matlab softwares is selected in described step 4 Carry out multiple dimensioned discrete wavelet transformation, Selection of Wavelet Basis bior2.2 biorthogonal wavelets, multi-scale wavelet decomposition to the third level.
- A kind of 9. Remote Sensing Image Texture analysis side based on multi-scale wavelet decomposition and fractal theory according to claim 8 Method, it is characterised in that：Distinguished in described step 5 by FRACLAB softwares in calculation procedure 4 the 1st, 2,3 grade after wavelet decomposition The box meter dimension values and multifractal spectra of small echo high frequency, low frequency signal.
- It is 10. according to claim 9 a kind of based on the analysis of the Remote Sensing Image Texture of multi-scale wavelet decomposition and fractal theory Method, it is characterised in that：Precision parameter is set as 0.40 during being calculated in described step 5, and dimension computational methods choose box Dimension is counted, area grayscale statistical method chooses foundation that is using selection area gray value and being calculated as dimension.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710738940.4A CN107609578A (en) | 2017-08-25 | 2017-08-25 | Remote sensing image texture analysis method based on multi-scale wavelet decomposition and fractal theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710738940.4A CN107609578A (en) | 2017-08-25 | 2017-08-25 | Remote sensing image texture analysis method based on multi-scale wavelet decomposition and fractal theory |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107609578A true CN107609578A (en) | 2018-01-19 |
Family
ID=61065917
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710738940.4A Pending CN107609578A (en) | 2017-08-25 | 2017-08-25 | Remote sensing image texture analysis method based on multi-scale wavelet decomposition and fractal theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107609578A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110147802A (en) * | 2019-05-13 | 2019-08-20 | 安徽工业大学 | The inertinite classification method and system of trend fluction analysis are gone based on multi-fractal |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101551851A (en) * | 2008-03-31 | 2009-10-07 | 中国科学院沈阳自动化研究所 | Infrared image target identification method |
-
2017
- 2017-08-25 CN CN201710738940.4A patent/CN107609578A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101551851A (en) * | 2008-03-31 | 2009-10-07 | 中国科学院沈阳自动化研究所 | Infrared image target identification method |
Non-Patent Citations (1)
| Title |
|---|
| 田青林: ""鞍本古陆块图像识别方法研究"", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110147802A (en) * | 2019-05-13 | 2019-08-20 | 安徽工业大学 | The inertinite classification method and system of trend fluction analysis are gone based on multi-fractal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ourhzif et al. | Lithological mapping using landsat 8 oli and aster multispectral data in imini-ounilla district south high atlas of marrakech | |
| Proisy et al. | Predicting and mapping mangrove biomass from canopy grain analysis using Fourier-based textural ordination of IKONOS images | |
| CN111242224B (en) | Multi-source remote sensing data classification method based on unmanned aerial vehicle extraction classification sample points | |
| Serkan Öztan et al. | Mapping evaporate minerals by ASTER | |
| CN107491758B (en) | Yangtze river basin water body information extraction and space coding method thereof | |
| CN113240758B (en) | Remote sensing image fusion method, system, equipment and medium based on fusion derivative index | |
| Chen et al. | Digital mapping of the soil thickness of loess deposits over a calcareous bedrock in central France | |
| Cheng et al. | A fractal concentration–area method for assigning a color palette for image representation | |
| CN113222054A (en) | Remote sensing image fusion method, system, equipment and medium based on characteristic ratio index | |
| CN113191745B (en) | Remote sensing large-range evaluation method and medium for real estate construction project construction progress | |
| CN107504923A (en) | A kind of comprehensive remote sensing image and the laminaria culture area monitoring method for prolonging rope information | |
| CN113255144B (en) | Shallow sea remote sensing water depth inversion method based on FUI partition and Randac | |
| CN109598202A (en) | A kind of object-based satellite image multi objective built-up areas extraction method | |
| CN110532615A (en) | A kind of decomposition method step by step of shallow sea complicated landform | |
| CN104134069A (en) | Automatic identification method for shale microsections | |
| Drury et al. | Display and enhancement of gridded aeromagnetic data of the Solway Basin | |
| CN107609578A (en) | Remote sensing image texture analysis method based on multi-scale wavelet decomposition and fractal theory | |
| CN108491869A (en) | A kind of principal component transform remote sensing image fusion method that panchromatic wave-band gray value adaptively inverts | |
| CN117935079A (en) | Remote sensing image fusion method, system and readable storage medium | |
| CN116091850B (en) | Mining area land coverage classification model establishment and classification method | |
| Laake | Structural mapping with spectral attributes | |
| CN111160182B (en) | Method for identifying ancient river ground surface calcium rock formation | |
| Krahtopoulou et al. | Late Pleistocene-Holocene shoreline reconstruction and human exploitation of molluscan resources in northern Pieria, Macedonia, Greece | |
| CN114612578A (en) | Method and device for identifying effective reservoir | |
| CN111160184A (en) | Method for identifying ancient river channel in grassland coverage area |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180119 |
|
| RJ01 | Rejection of invention patent application after publication |