CN108280419A - Spatial feature detection method and system - Google Patents
Spatial feature detection method and system Download PDFInfo
- Publication number
- CN108280419A CN108280419A CN201810048907.3A CN201810048907A CN108280419A CN 108280419 A CN108280419 A CN 108280419A CN 201810048907 A CN201810048907 A CN 201810048907A CN 108280419 A CN108280419 A CN 108280419A
- Authority
- CN
- China
- Prior art keywords
- space
- vector
- pixel
- reflectivity
- curve
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/58—Extraction of image or video features relating to hyperspectral data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a spatial feature detection method and a spatial feature detection system. The method comprises the following steps: obtaining a remote sensing image of a target object, wherein the remote sensing image comprises a plurality of pixel points; extracting a spectral curve of each pixel point from the remote sensing image; the abscissa of the spectral curve represents the wavelength of the pixel point and the ordinate of the spectral curve represents the reflectivity of the pixel point; constructing a space vector of each corresponding pixel point according to each spectral curve; constructing a vector cloud space according to the space vector of each pixel point; and calculating the characteristics of the vector cloud space through the co-occurrence matrix and the difference matrix, wherein the characteristics of the vector cloud space are the characteristics of the target object. The invention defines the space characteristic from two aspects of direction and linear characteristic, thereby accurately extracting the characteristic of the target object.
Description
Technical field
The present invention relates to space detection fields, more particularly to a kind of space characteristics detection method and system.
Background technology
Space detection technique is usually used in terrain classification, target detection etc..Common space detection technique has neighborhood inspection
Survey technology, linear Feature Extraction Technology etc..Existing space detection technique pertains only to each put in image reflectance value
Neighborhood problem.Especially for the extraction of the linear feature in space, such as geologic prospect, highway detection, pipe detection
Etc. the detection of dangerous signs or phenomenons feature or the common feature extraction such as Texture classification seem not as one wishes.And since geological structure is reflected in remote sensing
Form is polygon on image, such as big construction reflection linear structure width is very big, and the linear structure width of little structure reflection is very
It is small, therefore construct spatial structure characteristic unobvious.In addition, these linear features are also easy the spy with structure as bridge, dam etc.
Sign is mutually obscured, therefore structured place usually occurs and can not extract linear feature, does not have structured place to be extracted line instead
Shape feature, and linear feature is various to select.
Invention content
The object of the present invention is to provide a kind of space characteristics detection method and systems, in terms of direction and linear feature two
Space characteristics are limited, to accurately extract the feature of target object.
To achieve the above object, the present invention provides following schemes：
A kind of space characteristics detection method, the method includes：
The remote sensing images of target object are obtained, the remote sensing images include multiple pixels；
The curve of spectrum where extracting each pixel in the remote sensing images；The abscissa table of the curve of spectrum
The ordinate of the wavelength and the curve of spectrum that show pixel indicates the reflectivity of pixel；
The space vector of corresponding each pixel is built according to each curve of spectrum；
Vector cloud space is built according to the space vector of each pixel；
The feature in vector cloud space, the feature in vector cloud space are calculated by co-occurrence matrix and difference matrix
For the feature of target object.
Optionally, the space vector that corresponding each pixel is built according to each curve of spectrum, specifically includes：
Calculate the reflectivity mean value of all pixels point on the curve of spectrum：
Wherein, i=1,2 .., n, n are the number of pixel number point, and λ is the wavelength of pixel；fi(λ) indicates the curve of spectrum
The reflectivity of upper ith pixel point；
Calculate the difference of each pixel reflectivity of each wave band and the reflectivity mean value on the curve of spectrum：
Zim' be ith pixel point reflectivity of each wave band and the reflectivity mean value on the curve of spectrum difference；
ZimFor the reflectivity of ith pixel m-th of wave band of point；M=1,2 .., M；M is the number of wave band；
The space of each pixel is determined in the reflectivity of each wave band and the difference of the reflectivity mean value according to each pixel
Vector, formula are as follows：
For the space vector of ith pixel point.
Optionally, the feature that vector cloud space is calculated by co-occurrence matrix and difference matrix, specifically includes：
Coordinate mapping conversion is carried out to the remote sensing images according to the space vector of each pixel, after being converted
Image；
Obtain the histogram multi-dimensional matrix of the image after conversion；
The characteristic vector in vector cloud space is counted according to the histogram multi-dimensional matrix；
Vector cloud space is divided according to the characteristic vector, obtains multiple division regions；
Each space characteristics for dividing region are calculated by co-occurrence matrix and difference matrix.
Optionally, the space characteristics detection method further includes：It is calculated by co-occurrence matrix and difference matrix described
After the feature in vector cloud space,
The wave band of the remote sensing images is chosen using optimum index method, obtains choosing wave band；
The selection wave band is synthesized, false color image is obtained；
The space characteristics are optimized, the space characteristics after being optimized；
The space characteristics after the optimization are added in the false color image by the grid and vector addition method, are obtained most
Whole image；The final image includes the space characteristics after optimization.
Optionally, the space characteristics are optimized by the auto-correlation function in vector cloud space.
The present invention also provides a kind of space characteristics detecting system, the system comprises：
Acquisition module, the remote sensing images for obtaining target object, the remote sensing images include multiple pixels；
Extraction module, for the curve of spectrum where extracting each pixel in the remote sensing images；The spectrum
The abscissa of curve indicates that the wavelength of pixel and the ordinate of the curve of spectrum indicate the reflectivity of pixel；
First structure module, the space vector for building corresponding each pixel according to each curve of spectrum；
Second structure module, for building vector cloud space according to the space vector of each pixel；
Computing module, the feature for calculating vector cloud space by co-occurrence matrix and difference matrix, the arrow
Amount cloud space is characterized as the feature of target object.
Optionally, the first structure module includes：
Average calculation unit, the reflectivity mean value for calculating all pixels point on the curve of spectrum, calculation formula is such as
Under：
Wherein, i=1,2 .., n, n are the number of pixel number point, and λ is the wavelength of pixel；fi(λ) indicates the curve of spectrum
The reflectivity of upper ith pixel point；
Difference computational unit, for calculating each pixel reflectivity of each wave band and reflection on the curve of spectrum
The difference of rate mean value, calculation formula are as follows：
Zim' be ith pixel point reflectivity of each wave band and the reflectivity mean value on the curve of spectrum difference；
ZimFor the reflectivity of ith pixel m-th of wave band of point；M=1,2 .., M；M is the number of wave band；
Determination unit, for being determined respectively in the reflectivity of each wave band and the difference of the reflectivity mean value according to each pixel
The space vector of pixel, formula are as follows：
For the space vector of ith pixel point.
Optionally, the computing module specifically includes：
Conversion unit turns for carrying out coordinate mapping to the remote sensing images according to the space vector of each pixel
Change, the image after being converted；
Second acquisition unit, the histogram multi-dimensional matrix for obtaining the image after converting；
Statistic unit, the characteristic vector for counting vector cloud space according to the histogram multi-dimensional matrix；
Division unit obtains multiple dividing regions for being divided to vector cloud space according to the characteristic vector
Domain；
Computing unit, for calculating each space characteristics for dividing region by co-occurrence matrix and difference matrix.
Optionally, the system also includes：
Module is chosen, for choosing the wave band of the remote sensing images using optimum index method, obtains choosing wave band；
Synthesis module obtains false color image for being synthesized to the selection wave band；
Optimization module, for being optimized to the space characteristics, the space characteristics after being optimized；
Laminating module, for the space characteristics after the optimization to be added to the pseudo color coding hologram by the grid and vector addition method
In image, final image is obtained；The final image includes the space characteristics after optimization.
Optionally, the optimization module carries out the space characteristics by the auto-correlation function in vector cloud space excellent
Change.
According to specific embodiment provided by the invention, the invention discloses following technique effects：According to each curve of spectrum structure
Build the space vector of corresponding each pixel；Vector cloud space is built according to the space vector of each pixel；Pass through symbiosis
Matrix and difference matrix calculate the feature in vector cloud space, and vector cloud space is characterized as the spy of target object
Sign.The present invention constructs a kind of vector cloud space, and space characteristics are limited in terms of direction and linear feature two, so as to
More accurately extract desired spatial information.Vector cloud space can carry out All Layers or several figure layers to obtain operation,
Meet different application.
Description of the drawings
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is a kind of flow chart of space characteristics detection method provided in an embodiment of the present invention；
Fig. 2 is the spy provided in an embodiment of the present invention that vector cloud space is calculated by co-occurrence matrix and difference matrix
The method flow diagram of sign；
Fig. 3 is a kind of structure diagram of space characteristics detecting system provided in an embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of space characteristics detection method and systems, in terms of direction and linear feature two
Space characteristics are limited, to accurately extract the feature of target object.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
As shown in Figure 1, detection method includes the following steps for a kind of space characteristics：
Step 101：The remote sensing images of target object are obtained, the remote sensing images include multiple pixels.
Step 102：The curve of spectrum where extracting each pixel in the remote sensing images；The curve of spectrum
Abscissa indicates that the wavelength of pixel and the ordinate of the curve of spectrum indicate the reflectivity of pixel.
Specifically, remote sensing images used are usually to be made of multiple wave bands, such as ETM has 6 visible light-shortwaves
Infrared band, ASTER have 9 visible light-short infrared wave bands etc..Each point can extract a spectrum on image
Curve, the method for use are exactly to extract light to combination with wavelength by each band value in remote sensing images preferred coordinates (x, y)
Spectral curve.
Step 103：The space vector of corresponding each pixel is built according to each curve of spectrum.
Specifically, calculating the reflectivity mean value of all pixels point on the curve of spectrum：
Wherein, i=1,2 .., n, n are the number of pixel number point, and λ is the wavelength of pixel；fi(λ) indicates the curve of spectrum
The reflectivity of upper ith pixel point；
Calculate the difference of each pixel reflectivity of each wave band and the reflectivity mean value on the curve of spectrum：
Zim' be ith pixel point reflectivity of each wave band and the reflectivity mean value on the curve of spectrum difference；
ZimFor the reflectivity of ith pixel m-th of wave band of point；M=1,2 .., M；M is the number of wave band；
The space of each pixel is determined in the reflectivity of each wave band and the difference of the reflectivity mean value according to each pixel
Vector, formula are as follows：
For the space vector of ith pixel point.
Step 104：Vector cloud space is built according to the space vector of each pixel.
Step 105：The feature in vector cloud space is calculated by co-occurrence matrix and difference matrix, the vector cloud is empty
Between be characterized as the feature of target object.Specifically, as shown in Fig. 2, including the following steps：
Step 1051：Coordinate mapping conversion is carried out to the remote sensing images according to the space vector of each pixel, is obtained
Image after to conversion.
Specifically, by the processing to whole image, we obtain the spectral space vectors of each point.By each
A point all there is the coordinate value of oneself can be indicated using following representation method if being indicated each point using coordinate
For coordinate (xi,yi) space vector, Zi′1(xi1,yi1) it is coordinate (xi1,yi1) first band DN
Value or reflectivity.If being entered as new image to each coordinate of pressing, entire remote sensing images are all converted,
So image is converted to spectrum arrow form
Step 1052：Obtain the histogram multi-dimensional matrix of the image after conversion.
Step 1053：The characteristic vector in vector cloud space is counted according to the histogram multi-dimensional matrix.
Step 1054：Vector cloud space is divided according to the characteristic vector, obtains multiple division regions.
By the aspect ratio pair of Statistical Vector, to be determined distance and the direction of space characteristics, according to space characteristics three
Kurtosis, variance, degree of bias direction vector and mould that matrix is established are tieed up, the statistics to determine rough image space feature (texture) is special
Sign, the size of active window is primarily determined according to these features.
The mould of kurtosis vector is bigger than normal, illustrates that at least one layer of histogram is more precipitous, and space characteristics primitive window is suitble to small window
Mouthful, if degree of bias mould is bigger than normal, illustrate at least one layer of histogram map migration, positively biased represents that light tone pixel is more, and negative bias represents dead color pixel
It is more, then space characteristics primitive window is suitble to wicket to illustrate that at least one layer of histogram is more flat if variance mould is big,
Space characteristics primitive window is suitble to compared with big window.Space vector is biased to that axis, illustrates that this axis characteristic parameter is larger.Kurtosis
Space characteristics correlation can be characterized with the angle of the degree of bias and variance vectors, is determined by the parameter of these angles suitable
Window, these features are also the feature that the rectangle to be formed also illustrates that space.
Step 1055：Each space characteristics for dividing region are calculated by co-occurrence matrix and difference matrix.
Specifically, the window obtained according to space characteristics determines primitive window in the window of vector cloud spatial choice primitive
The direction of interior texture and distance calculate its multidimensional co-occurrence matrix and difference matrix.
Take a vector in vector cloud spaceWith another arrow along certain orientation and offset distance offset
Measure A1(x+i, y+j) obtains the matrix [A, A1], it is moved in space vector cloud if calculated, calculates various matrix [Ai,Aj],
Statistical matrix [Ai,Aj] prospects numberDivided by total degree R, prospects probability density is obtained, vector co-occurrence matrix is formed
Entropy, inertia, energy are calculated by co-occurrence matrix；
Entropy
Inertia
Energy
Take a vector in vector cloud spaceWith another arrow along certain orientation and offset distance offset
Measure A1(x+i, y+j) calculates its space angle θ, is moved in space vector cloud if calculated, calculates various angle (θi,θj),
Count the number P (θ of angle prospectsi,θj) divided by total degree R, prospects probability density is obtained, vector difference matrix δ is formedi,j
(θi,θj)。
Curvature, tonicity, enthalpy are calculated by calculating difference matrix；
Curvature
Tonicity
Enthalpy
According to specific embodiment provided by the invention, the invention discloses following technique effects：According to each curve of spectrum structure
Build the space vector of corresponding each pixel；Vector cloud space is built according to the space vector of each pixel；Pass through symbiosis
Matrix and difference matrix calculate the feature in vector cloud space, and vector cloud space is characterized as the spy of target object
Sign.The present invention constructs a kind of vector cloud space, and space characteristics are limited in terms of direction and linear feature two, so as to
More accurately extract desired spatial information.Vector cloud space can carry out All Layers or several figure layers to obtain operation,
Meet different application.
The method further includes：
The wave band of the remote sensing images is chosen using optimum index method, obtains choosing wave band.
The selection wave band is synthesized, false color image is obtained.
Specifically, false color image synthesis is synthesized using tri- wave bands of RGB, it is larger that synthesis carries out selection information content
Wave band is synthesized.The selection of wave band is carried out using optimum index method：
In formula, SiFor the standard deviation for i-th of wave band, RI, jFor the related coefficient of i-th, j wave bands.OIF is bigger, shows to wrap
The information content contained is bigger, and therefore, maximum OIF band combinations are optimal bands combined, the figure synthesized using optimal bands combined
As being used as base map.
The space characteristics are optimized, the space characteristics after being optimized.
The optimization of space characteristics is carried out using vector cloud autocorrelation function C (ε, η, i, j, k), characterization space characteristics are thick
Rugosity：
In formula, i, j are coordinate, and limitation parameter ω arranges window, and ε, η are the offset of pixel, and f is function.It is coarse
Then C (ε, η, i, j, k) correlation is high, can indicate roughness with matrix T；
The space characteristics after the optimization are added in the false color image by the grid and vector addition method, are obtained most
Whole image；The final image includes the space characteristics after optimization.
The pseudo color coding hologram figure with the maximum band combination of comentropy, vector is used to use with same projection in base map
Point-line-surface indicates.It is layered using coordinate and grid and vector is overlapped processing.To form the image for being suitble to human eye custom.
Vector f (x, y, z), x, y are corresponding coordinate, and z is characterized value, and f (x, y, z) is vector value, grid WithFor corresponding coordinate,For grid gray value.
It enables,To realize gridGray value and vectorSuperposition.
As shown in figure 3, the present invention also provides a kind of space characteristics detecting systems.The system comprises：
Acquisition module 301, the remote sensing images for obtaining target object, the remote sensing images include multiple pixels；
Extraction module 302, for the curve of spectrum where extracting each pixel in the remote sensing images；The light
The abscissa of spectral curve indicates that the wavelength of pixel and the ordinate of the curve of spectrum indicate the reflectivity of pixel.
First structure module 303, the space vector for building corresponding each pixel according to each curve of spectrum.
The first structure module 303 specifically includes：
Average calculation unit, the reflectivity mean value for calculating all pixels point on the curve of spectrum, calculation formula is such as
Under：
Wherein, i=1,2 .., n, n are the number of pixel number point, and λ is the wavelength of pixel；fi(λ) indicates the curve of spectrum
The reflectivity of upper ith pixel point；
Difference computational unit, for calculating each pixel reflectivity of each wave band and reflection on the curve of spectrum
The difference of rate mean value, calculation formula are as follows：
Zim' be ith pixel point reflectivity of each wave band and the reflectivity mean value on the curve of spectrum difference；
ZimFor the reflectivity of ith pixel m-th of wave band of point；M=1,2 .., M；M is the number of wave band；
Determination unit, for being determined respectively in the reflectivity of each wave band and the difference of the reflectivity mean value according to each pixel
The space vector of pixel, formula are as follows：
For the space vector of ith pixel point.
Second structure module 304, for building vector cloud space according to the space vector of each pixel.
Computing module 305, the feature for calculating vector cloud space by co-occurrence matrix and difference matrix are described
Vector cloud space is characterized as the feature of target object.
The computing module 305 specifically includes：
Conversion unit turns for carrying out coordinate mapping to the remote sensing images according to the space vector of each pixel
Change, the image after being converted；
Second acquisition unit, the histogram multi-dimensional matrix for obtaining the image after converting；
Statistic unit, the characteristic vector for counting vector cloud space according to the histogram multi-dimensional matrix；
Division unit obtains multiple dividing regions for being divided to vector cloud space according to the characteristic vector
Domain；
Computing unit, for calculating each space characteristics for dividing region by co-occurrence matrix and difference matrix.
The system also includes：
Module is chosen, for choosing the wave band of the remote sensing images using optimum index method, obtains choosing wave band.
Synthesis module obtains false color image for being synthesized to the selection wave band.
Optimization module, for being optimized to the space characteristics, the space characteristics after being optimized；The optimization module
The space characteristics are optimized by the auto-correlation function in vector cloud space.
Laminating module, for the space characteristics after the optimization to be added to the pseudo color coding hologram by the grid and vector addition method
In image, final image is obtained；The final image includes the space characteristics after optimization.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related place is said referring to method part
It is bright.
Principle and implementation of the present invention are described for specific case used herein, and above example is said
The bright method and its core concept for being merely used to help understand the present invention；Meanwhile for those of ordinary skill in the art, foundation
The thought of the present invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (10)
1. a kind of space characteristics detection method, which is characterized in that the method includes：
The remote sensing images of target object are obtained, the remote sensing images include multiple pixels；
The curve of spectrum where extracting each pixel in the remote sensing images；The abscissa of the curve of spectrum indicates picture
The ordinate of the wavelength of vegetarian refreshments and the curve of spectrum indicates the reflectivity of pixel；
The space vector of corresponding each pixel is built according to each curve of spectrum；
Vector cloud space is built according to the space vector of each pixel；
The feature in vector cloud space is calculated by co-occurrence matrix and difference matrix, vector cloud space is characterized as mesh
Mark the feature of object.
2. according to the method described in claim 1, it is characterized in that, described build corresponding each pixel according to each curve of spectrum
Space vector, specifically include：
Calculate the reflectivity mean value of all pixels point on the curve of spectrum：
Wherein, i=1,2 .., n, n are the number of pixel number point, and λ is the wavelength of pixel；fi(λ) is indicated i-th on the curve of spectrum
The reflectivity of a pixel；
Calculate the difference of each pixel reflectivity of each wave band and the reflectivity mean value on the curve of spectrum：
Zim' be ith pixel point reflectivity of each wave band and the reflectivity mean value on the curve of spectrum difference；ZimFor
The reflectivity of ith pixel m-th of wave band of point；M=1,2 .., M；M is the number of wave band；
The space vector of each pixel is determined in the reflectivity of each wave band and the difference of the reflectivity mean value according to each pixel,
Formula is as follows：
For the space vector of ith pixel point.
3. according to the method described in claim 1, it is characterized in that, described by described in co-occurrence matrix and difference matrix calculating
The feature in vector cloud space, specifically includes：
Coordinate mapping conversion, the figure after being converted are carried out to the remote sensing images according to the space vector of each pixel
Picture；
Obtain the histogram multi-dimensional matrix of the image after conversion；
The characteristic vector in vector cloud space is counted according to the histogram multi-dimensional matrix；
Vector cloud space is divided according to the characteristic vector, obtains multiple division regions；
Each space characteristics for dividing region are calculated by co-occurrence matrix and difference matrix.
4. according to the method described in claim 3, it is characterized in that, the space characteristics detection method further includes：Described logical
Cross co-occurrence matrix and after difference matrix calculates the feature in vector cloud space,
The wave band of the remote sensing images is chosen using optimum index method, obtains choosing wave band；
The selection wave band is synthesized, false color image is obtained；
The space characteristics are optimized, the space characteristics after being optimized；
The space characteristics after the optimization are added in the false color image by the grid and vector addition method, are finally schemed
Picture；The final image includes the space characteristics after optimization.
5. according to the method described in claim 4, it is characterized in that, by the auto-correlation function in vector cloud space to described
Space characteristics optimize.
6. a kind of space characteristics detecting system, which is characterized in that the system comprises：
Acquisition module, the remote sensing images for obtaining target object, the remote sensing images include multiple pixels；
Extraction module, for the curve of spectrum where extracting each pixel in the remote sensing images；The curve of spectrum
Abscissa indicate pixel wavelength and the curve of spectrum ordinate indicate pixel reflectivity；
First structure module, the space vector for building corresponding each pixel according to each curve of spectrum；
Second structure module, for building vector cloud space according to the space vector of each pixel；
Computing module, the feature for calculating vector cloud space by co-occurrence matrix and difference matrix, the vector cloud
Space is characterized as the feature of target object.
7. system according to claim 6, which is characterized in that described first, which builds module, includes：
Average calculation unit, the reflectivity mean value for calculating all pixels point on the curve of spectrum, calculation formula are as follows：
Wherein, i=1,2 .., n, n are the number of pixel number point, and λ is the wavelength of pixel；fi(λ) is indicated i-th on the curve of spectrum
The reflectivity of a pixel；
Difference computational unit, it is equal for calculating each pixel reflectivity of each wave band and reflectivity on the curve of spectrum
The difference of value, calculation formula are as follows：
Zim' be ith pixel point reflectivity of each wave band and the reflectivity mean value on the curve of spectrum difference；ZimFor
The reflectivity of ith pixel m-th of wave band of point；M=1,2 .., M；M is the number of wave band；
Determination unit, for determining each pixel in the reflectivity of each wave band and the difference of the reflectivity mean value according to each pixel
The space vector of point, formula are as follows：
For the space vector of ith pixel point.
8. system according to claim 6, which is characterized in that the computing module specifically includes：
Conversion unit is obtained for carrying out coordinate mapping conversion to the remote sensing images according to the space vector of each pixel
Image after to conversion；
Second acquisition unit, the histogram multi-dimensional matrix for obtaining the image after converting；
Statistic unit, the characteristic vector for counting vector cloud space according to the histogram multi-dimensional matrix；
Division unit obtains multiple division regions for being divided to vector cloud space according to the characteristic vector；
Computing unit, for calculating each space characteristics for dividing region by co-occurrence matrix and difference matrix.
9. system according to claim 8, which is characterized in that the system also includes：
Module is chosen, for choosing the wave band of the remote sensing images using optimum index method, obtains choosing wave band；
Synthesis module obtains false color image for being synthesized to the selection wave band；
Optimization module, for being optimized to the space characteristics, the space characteristics after being optimized；
Laminating module, for the space characteristics after the optimization to be added to the false color image by the grid and vector addition method
In, obtain final image；The final image includes the space characteristics after optimization.
10. system according to claim 9, which is characterized in that the optimization module by vector cloud space from
Correlation function optimizes the space characteristics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810048907.3A CN108280419B (en) | 2018-01-18 | 2018-01-18 | Spatial feature detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810048907.3A CN108280419B (en) | 2018-01-18 | 2018-01-18 | Spatial feature detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108280419A true CN108280419A (en) | 2018-07-13 |
| CN108280419B CN108280419B (en) | 2020-06-30 |
Family
ID=62804096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810048907.3A Active CN108280419B (en) | 2018-01-18 | 2018-01-18 | Spatial feature detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108280419B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597027A (en) * | 2018-12-06 | 2019-04-09 | 清华大学 | A kind of positioning system and method based on single base station |
| CN110825109A (en) * | 2019-11-12 | 2020-02-21 | 广州极飞科技有限公司 | Spectrum information acquisition method and device and aircraft control method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130129208A1 (en) * | 2011-11-21 | 2013-05-23 | Tandent Vision Science, Inc. | Color analytics for a digital image |
| CN103576132A (en) * | 2012-07-20 | 2014-02-12 | 上海莱凯数码科技有限公司 | Processing method and system for remote-sensing images |
| CN103868858A (en) * | 2014-03-03 | 2014-06-18 | 上海交通大学 | Method for determining optimal band realizing spectrum response of dominated salt in saline soil |
| CN104122210A (en) * | 2014-07-02 | 2014-10-29 | 中国林业科学研究院林业研究所 | Hyperspectral waveband extraction method based on optimum index factor-correlation coefficient method |
| CN107506769A (en) * | 2017-08-03 | 2017-12-22 | 深圳市盛路物联通讯技术有限公司 | A kind of extracting method and system of urban water-body information |
-
2018
- 2018-01-18 CN CN201810048907.3A patent/CN108280419B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130129208A1 (en) * | 2011-11-21 | 2013-05-23 | Tandent Vision Science, Inc. | Color analytics for a digital image |
| CN103576132A (en) * | 2012-07-20 | 2014-02-12 | 上海莱凯数码科技有限公司 | Processing method and system for remote-sensing images |
| CN103868858A (en) * | 2014-03-03 | 2014-06-18 | 上海交通大学 | Method for determining optimal band realizing spectrum response of dominated salt in saline soil |
| CN104122210A (en) * | 2014-07-02 | 2014-10-29 | 中国林业科学研究院林业研究所 | Hyperspectral waveband extraction method based on optimum index factor-correlation coefficient method |
| CN107506769A (en) * | 2017-08-03 | 2017-12-22 | 深圳市盛路物联通讯技术有限公司 | A kind of extracting method and system of urban water-body information |
Non-Patent Citations (5)
| Title |
|---|
| RIAD KHELIFI 等: "Texture Classification For Multi-Spectral Images Using Spatial And Spectral Gray Level Differences", 《IMAGE PROCESSING THEORY, TOOLS AND APPLICATIONS》 * |
| YUANYUAN CHEN 等: "A Spectral Signature Shape-Based Algorithm for Landsat Image Classification", 《NTERNATIONAL JOURNAL OF GEO-INFORMATION》 * |
| YU-JUN ZHANG 等: "Interpreting the Shortwave Infrared & Thermal Infrared Regions of Remote Sensed Electromagnetic Spectrum with Application for Mineral-Deposits Exploration", 《JOURNAL OF APPLIED MATHEMATICS AND PHYSICS》 * |
| 张玉君 等: "利用光谱角填图（SAM）优化多光谱遥感异常", 《矿物学报 增刊》 * |
| 骆剑承 等: "支撑向量机及其遥感影像空间特征提取和分类的应用研究", 《遥感学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597027A (en) * | 2018-12-06 | 2019-04-09 | 清华大学 | A kind of positioning system and method based on single base station |
| CN109597027B (en) * | 2018-12-06 | 2021-08-17 | 清华大学 | Positioning system and method based on single base station |
| CN110825109A (en) * | 2019-11-12 | 2020-02-21 | 广州极飞科技有限公司 | Spectrum information acquisition method and device and aircraft control method |
| CN110825109B (en) * | 2019-11-12 | 2023-11-21 | 广州极飞科技股份有限公司 | Method and device for acquiring spectrum information and method for controlling aircraft |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108280419B (en) | 2020-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104574347B (en) | Satellite in orbit image geometry positioning accuracy evaluation method based on multi- source Remote Sensing Data data | |
| CN101512601B (en) | Method for determining a depth map from images, device for determining a depth map | |
| CN106600574B (en) | A kind of landslide extracting method based on remote sensing image and altitude data | |
| CN105651263B (en) | Shallow water depth multi-source remote sensing merges inversion method | |
| CN106408513B (en) | Depth map super resolution ratio reconstruction method | |
| CN104361590B (en) | High-resolution remote sensing image registration method with control points distributed in adaptive manner | |
| CN104952107A (en) | Three-dimensional bridge reconstruction method based on vehicle-mounted LiDAR point cloud data | |
| CN103489193A (en) | High-resolution remote-sensing image change detection method facing targets and based on integrating strategy | |
| CN108280419A (en) | Spatial feature detection method and system | |
| CN110991738A (en) | NPP/VIIRS night lamplight remote sensing data-based economic development research method | |
| CN105469392B (en) | High spectrum image conspicuousness detection method based on the comparison of region spectrum Gradient Features | |
| CN113916130A (en) | Building position measuring method based on least square method | |
| CN110276746A (en) | A kind of robustness method for detecting change of remote sensing image | |
| CN113191213A (en) | High-resolution remote sensing image newly-added building detection method | |
| CN102878985B (en) | Water surface wave scale monitoring method based on image texture features | |
| CN114265062B (en) | InSAR phase unwrapping method based on phase gradient estimation network | |
| CN116468769A (en) | Depth information estimation method based on image | |
| CN114170526A (en) | Remote sensing image multi-scale target detection and identification method based on lightweight network | |
| CN102938148A (en) | High-spectrum image texture analysis method based on V-GLCM (Gray Level Co-occurrence Matrix) | |
| CN112132950A (en) | Three-dimensional point cloud scene updating method based on crowdsourcing image | |
| CN116778341A (en) | Multi-view feature extraction and identification method for radar image | |
| CN109829939A (en) | A method of it reducing multi-view images and matches corresponding image points search range | |
| CN110276240A (en) | A kind of SAR image building wall window information extracting method | |
| CN109448122A (en) | The method and application of deformation measurement are carried out to the discontinuous body comprising structural plane | |
| CN116958800A (en) | Remote sensing image change detection method based on hierarchical attention residual unet++ |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |