CN107274421A - A kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment - Google Patents
A kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment Download PDFInfo
- Publication number
- CN107274421A CN107274421A CN201710301493.6A CN201710301493A CN107274421A CN 107274421 A CN107274421 A CN 107274421A CN 201710301493 A CN201710301493 A CN 201710301493A CN 107274421 A CN107274421 A CN 107274421A
- Authority
- CN
- China
- Prior art keywords
- pixel
- spatial domain
- remote sensing
- codomain
- vector
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
Abstract
When the present invention provides a kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment and is used to solving the scale parameter that calculates in the prior art for primitive in segmentation and physics image, the problem of the precision and not high efficiency of segmentation.Wherein remote sensing image dimension calculation method, including：S101 obtains vector marginal information；S102 is based on vector marginal information and calculates spatial domain yardstick, obtains spatial domain scalogram picture；S103 is based on vector marginal information and spatial domain scalogram picture calculates and obtains codomain scalogram picture.Vector edge and its shape facility information in remote sensing image as constraints, is carried out the spatial domain yardstick and codomain dimension calculation of Remote Sensing Image Segmentation by the present invention.The scale parameter of multi-scale segmentation of remote sensing images is obtained in terms of geographical space and spectral space two respectively, and improves the precision and efficiency of Image Segmentation.
Description
Technical field
The present invention relates to computer technology, and in particular to a kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and
Computer equipment.
Background technology
Based on object remote sensing images analysis research field, dimension calculation is an important research direction.At present, surround
Multiscale simulation method perhaps has been proposed in relevant remote sensing image image fusion segmentation.Due to high spatial resolution remote sense image
The complexity of data itself, and geographic object size and the otherness of Spatial Distribution Pattern, it is difficult to set up a global yardstick
Parameter model effectively instructs big region multi-scale division parameter (while carrying out adjacent pixel homogeneity in spatial domain and spectral domain
Property estimate with heterogeneity) calculating.Engineering practice shows, adaptively obtains essence of the rational scale parameter not only to segmentation result
True property plays a key effect, and profound influence effective identification and the post-processing of physics image primitive.
Yardstick is the scale for describing the heterogeneous threshold value of multi-scale segmentation of remote sensing images, inside measurement physics image primitive
The semantic silhouette target of pixel homogeneity and measure space atural object heterogeneous function to each other, rational yardstick causes physics image
Primitive farthest coincide with the adopted silhouette target of corresponding spatially story on edge.Over nearly 10 years, researcher is successive
Propose object-based remote sensing images analysis method and correlation technique is realized；As " high-resolution earth observation it is some before
Along problem in science " --- Remote sensing image understanding and the image fusion of one of information extraction core technology are split is given birth to any yardstick
The physics image primitive (the adjoining pixel set for meeting geometry and spectrum homogeney criterion) enriched into geometry and attribute information, after
And using physics image primitive as fundamental space analytic unit, story justice over the ground is realized using its geometric properties and spectrum statistical information
The automatic identification of silhouette target (the ground semantic description of physics image primitive or the imagery zone formed by its combination) and classification.
Multiscale morphology Image Segmentation is used as the extensive pass that researcher is obtained based on one of object remote sensing images analysis core technology
Note, be currently based on object remote sensing images analysis method and tentatively commercialized image fusion dividing method to be integrated inAnd FeatureAlgorithm in software is representative；In addition, the image point that all kinds of documents are introduced
Segmentation method and the algorithm of realization are a lot；Adjacent pixel is in spatial domain when appropriate scale parameter is to describe to split in these algorithms
With the heterogeneity and homogeney scope of spectral domain.But when these methods or algorithm are applied to high spatial resolution remote sense image segmentation
Obvious limitation is still suffered from, the limitation is presented as the scale parameter calculated in the prior art for segmentation physics image
In primitive when, the precision and efficiency of segmentation be not high.
The content of the invention
In view of the above problems, the present invention proposes the one kind for overcoming above mentioned problem or solving the above problems at least in part
Remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment.
For this purpose, in a first aspect, the present invention proposes a kind of remote sensing image dimension calculation method, including：
S101 obtains vector marginal information；
S102 is based on vector marginal information and calculates spatial domain yardstick, obtains spatial domain scalogram picture；
S103 is based on vector marginal information and spatial domain scalogram picture calculates and obtains codomain scalogram picture.
Optionally, include before step S101：
S201 is receiving remote sensing image；
S202 is pre-processed to remote sensing image, and pretreatment is included at panchromatic image fusion treatment and multispectral image filtering
Reason；
S203, to pretreated remote sensing image extract vector marginal information；
Optionally, the S102 is based on vector marginal information and calculates spatial domain yardstick, obtains spatial domain scalogram picture, including：
The distance that each pixel in remote sensing image arrives respective edges on k direction is calculated, pixel is in k direction
The upper distance to respective edges is respectively N1~Nk；
The pixel value S of the corresponding pixel of the pixel in the spatial domain yardstick image of settingdFor N1~NkAverage value.
Optionally, the S103 is based on vector marginal information and spatial domain scalogram picture calculates and obtains codomain scalogram picture, bag
Include：For each pixel in remote sensing image, calculate on eight directions from the pixel to corresponding edge, on each direction
The spectral value sum average value of all pixels point, obtains ME1~MEk；
Calculate ME1~MEkAverage value as the pixel codomain scale-value.
Optionally, the S103 is based on vector marginal information and spatial domain dimension calculation codomain yardstick, including：
For each pixel in the scalogram picture of spatial domain, calculate from the pixel to the K direction at corresponding edge,
The intermediate value of the spectral value of all pixels point, obtains MI on each direction1~MIk；
Calculate MI1~MIkIntermediate value as the pixel codomain scale-value.
Optionally, the S103 is based on vector marginal information and spatial domain dimension calculation codomain yardstick, including：
For each pixel in the scalogram picture of spatial domain, calculate on the K direction from the pixel to corresponding edge, often
The spectral value sum average value of all pixels point on individual direction, obtains ME1~MEk；
Calculate ME1~MEkIntermediate value as the pixel codomain scale-value.
Optionally, the multispectral image filtering process includes medium filtering, gaussian filtering, average drifting, bilateral filtering
Or the one or more in anisotropic diffusion filtering.
Second aspect, the present invention provides a kind of computer-readable recording medium, is stored thereon with computer program, the program
The step of as above any methods described being realized when being executed by processor.
Second aspect, the present invention provides a kind of computer equipment, including memory, processor and is stored in the storage
On device and the computer program that can perform on the processor, as above any institute is realized during the computing device described program
The step of stating method.
As shown from the above technical solution, the present invention using the vector edge and its shape facility information in remote sensing image as about
Beam condition, carries out the spatial domain yardstick and codomain dimension calculation of Remote Sensing Image Segmentation.Respectively from geographical space and the side of spectral space two
Face obtains the scale parameter of multi-scale segmentation of remote sensing images, and improves the precision and efficiency of Image Segmentation.
Above be to provide to the present invention some in terms of understanding simplified summary.This part neither the present invention and
The detailed statement of its various embodiment is nor the statement of exhaustion.It is both not used in identification the present invention important or key feature or
Do not limit the scope of the present invention, but the selected principle of the present invention provided with a kind of reduced form, as to it is given below more
The brief introduction specifically described.It should be appreciated that either alone or in combination using one for being set forth above or being detailed below or
Multiple features, other embodiments of the invention are also possible.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are the present invention
Some embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
These accompanying drawings obtain other accompanying drawings.
Fig. 1 is the method flow schematic diagram of one embodiment of the present of invention；
Fig. 2 a are eight direction schematic diagrams of a certain pixel in one embodiment of the present of invention；
Fig. 2 b be one embodiment of the present of invention in current pixel point to eight directions in edge distance；
Fig. 2 c are local optimum spatial domain scale parameter image Visualization example in one embodiment of the present of invention；
Fig. 3 a are the original remote sensing image in one embodiment of the present of invention；
Fig. 3 b are the corresponding codomain yardstick images of Fig. 3 a (image obtained by average-codomain scaling algorithm)；
Fig. 3 c are the corresponding codomain yardstick images of Fig. 3 a (image obtained by intermediate value-codomain scaling algorithm)；
Fig. 3 d are the corresponding codomain yardstick images of Fig. 3 a (image obtained by joint-codomain scaling algorithm)；
Fig. 4 a are the remote sensing image dimension calculation experimental data in one embodiment of the present of invention, wherein being labelled with to be identified
Region；
Fig. 4 b are remote sensing images of Fig. 4 a after gaussian filtering process；
Fig. 4 c are remote sensing images of Fig. 4 a after median filter process；
Fig. 4 d are remote sensing images of Fig. 4 a after mean filter is handled；
Fig. 4 e are remote sensing images of Fig. 4 a after bilateral filtering is handled；
Fig. 4 f be Fig. 4 a by) remote sensing image after average drifting filtering process；
Fig. 4 g are remote sensing images of Fig. 4 a after anisotropic diffusion filtering is handled；
Fig. 5 is the corresponding vector edge extraction results of Fig. 4 a；
Fig. 6 is the corresponding spatial domain yardstick images of Fig. 4 a；
Fig. 7 a are the corresponding codomain yardstick images of Fig. 4 a (image obtained by average-codomain scaling algorithm)；
Fig. 7 b are the corresponding codomain yardstick images of Fig. 4 a (image obtained by intermediate value-codomain scaling algorithm)；
Fig. 7 c are the corresponding codomain yardstick images of Fig. 4 a (image obtained by joint-codomain scaling algorithm).
Embodiment
The present invention is described below in conjunction with exemplary communication system.
It should be noted that such as S101, S102 etc used herein mark, its be not intended to imply S101 and
Order between S102, but for the ease of referring to specific step with such mark in the de-scription, have reference herein
Body understands, is easy to read.
As shown in figure 1, the present embodiment discloses a kind of remote sensing image dimension calculation method, including：
S101 obtains vector marginal information；
S102 is based on vector marginal information and calculates spatial domain yardstick, obtains spatial domain scalogram picture；
S103 is based on vector marginal information and spatial domain scalogram picture calculates and obtains codomain scalogram picture.
Spatial domain yardstick is calculated using information such as the Geometry edge features and shape facility of remote sensing image and is determined remote sensing image point
The homogeney imaged object size cut simultaneously determines spatial dimension of the optimal physical image primitive in spatial domain.
Codomain yardstick is on the basis of the dimension calculation of spatial domain, the statistics such as integrated use average value, intermediate value, variance, standard deviation
Amount, calculates the spectral information of pixel in physics image primitive, obtains the codomain yardstick of homogeney imaged object spectral domain.
In the open embodiment of the present invention, the remote sensing image of input is as shown in fig. 4 a, it is necessary to explanation, Fig. 4
In be manually labelled with the position of object to be identified.The vector marginal information that Fig. 4 a are obtained after treatment is as shown in figure 5, pass through
Fig. 5, which is calculated, obtains spatial domain yardstick, and the image for carrying out Visualization acquisition to spatial domain yardstick is as shown in Figure 6；Based on vector edge
Information and spatial domain scalogram picture, which are calculated, obtains codomain yardstick；The image that the corresponding codomain yardsticks of Fig. 4 a are obtained by Visualization
As shown in Fig. 7a-7c, image such as Fig. 3 b-3d that the corresponding codomain yardsticks of Fig. 3 a are obtained by Visualization.
Vector edge and its shape facility information in remote sensing image as constraints, is carried out remote sensing image by the present invention
The spatial domain yardstick of segmentation and codomain dimension calculation.Remote sensing image to be obtained in terms of geographical space and spectral space two multiple dimensioned respectively
The scale parameter of segmentation, and improve the precision and efficiency of Image Segmentation.
It is appreciated that vector marginal information can be obtained by following methods：
S201 is receiving remote sensing image；Remote sensing image is as shown in Figure 3 a.
S202 is pre-processed to remote sensing image, and pretreatment is included at panchromatic image fusion treatment and multispectral image filtering
Reason；
S203, to pretreated remote sensing image extract vector marginal information；
Panchromatic image fusion treatment includes GS fusions (Gram-Schmidt fusions).In one embodiment, GS fusions bag
Include：The first step, copies a panchromatic wave-band from the wave band of low resolution.Second step, to the panchromatic wave-band that copies and many
Wave band carries out Gram-Schmidt conversion, and wherein panchromatic wave-band is used as first wave band.3rd step, with high spatial resolution
Panchromatic wave-band replaces first wave band after Gram-Schmidt conversion.Finally, melted using Gram-Schmidt inverse transformations
Close image.
Multispectral image filtering process includes medium filtering, gaussian filtering, average drifting, bilateral filtering or anisotropy and expanded
Dissipate filtering etc..Multispectral image filtering process can be the one or more of above processing.For single band grayscale image,
Spectral information in multiband high spatial resolution remote sense image available for rim detection more enriches.High spatial resolution remote sense
Three standard wave bands of image RGB are in RGB, IHS, Y IQ, YUV, C IELUV color spaces to various atural object vector edges
Effective extraction of information.Due to the influence of spectral range difference, in above-mentioned color space during different type of ground objects rim detections
Responsiveness makes a marked difference, so as to select different type of ground objects rim detection features in high spatial resolution remote sense image
Select and extract.
In step S202, panchromatic image fusion is combined with beneficial to panchromatic image space with multispectral image filtering process
Mutual supplement with each other's advantages between resolution ratio and multispectral image spectral resolution.And implement filtering process and reach the same of enhancing edge feature
When eliminate image data in interference noise；
And carry out remote sensing image vector edge extraction using multiband vector edge detection method in step S203
And false edge information is rejected.
In one embodiment, the S102 is based on vector marginal information and calculates spatial domain yardstick, obtains spatial domain scalogram picture
Including：
The distance that each pixel in remote sensing image arrives respective edges on k direction is calculated, pixel is in k direction
The upper distance to respective edges is respectively N1~Nk；
The pixel value S of the corresponding pixel of the pixel in the spatial domain yardstick image of settingdFor N1~NkAverage value.K etc.
In 2 n powers, n is more than or equal to 2；That is K can be 4,8,16 etc..
In one embodiment of the invention, each pixel calculated in remote sensing image arrives phase in 8 directions
The distance (i.e. K=8) at edge is answered, the distance that pixel arrives respective edges in 8 directions is respectively N1~N8, including：To view picture
Remote sensing image (original remote sensing image as shown in fig. 4 a) is arranged line by line to be handled pixel-by-pixel.By eight directions of current pixel point
(as shown in Figure 2 a, eight directions are respectively 0 degree counterclockwise, 45 degree, 90 degree, 135 degree, 180 degree, 225 degree, 270 degree and 360 degree)
To vector Edge Distance (i.e. N1~N8) and average value as this pixel new pixel value；If it is understood that K=4
(it is respectively then N in 4 directions (4 directions divide 360 degree equally) to vector Edge Distance by current pixel point1-N4), if
It is that K=16 is then 360 degree of current pixel 16 directions produced after dividing equally by 16 deciles, common property life N1-N116Totally 16 distances
Value.
Using the new pixel value as the corresponding spatial domain scale-value of the pixel, i.e., pixel correspondence in spatial domain yardstick image
Pixel pixel value SdFor the new pixel value, i.e.,：.
Wherein N=8, i=1,2 ..., 8；
To carrying out what image Visualization was obtained according to the local optimum spatial domain scale parameter data that new pixel value is generated
Image is as shown in Fig. 2 c, 2b.
Remote sensing image codomain dimension calculation under the conditions of the dimensional constraints of spatial domain has three kinds of methods, average-codomain two time scales approach,
Intermediate value-codomain two time scales approach, joint-codomain two time scales approach (also referred to as average-intermediate value-codomain two time scales approach).For ease of to three
The method of kind is described, and first two basic terms are defined.
First is mean direction pixel value, and referring to pixel, (edge is by vector marginal information to semantic silhouette target edge
Represent, object edge refers to specific pixel and extends first edge running into along specific direction) eight directions on, Mei Gefang
The average value of upward all pixels spectral value sum, is represented with ME.Calculation formula is as follows, and eight directions one have eight ME
Value, is designated as ME respectively1~ME8。
Wherein i represents direction, and one has 8 directions, i=1,2 ..., 8；N is the number of pixel on each direction, j=
1,2 ..., n.PjIt is pixel j spectral value.
Second is Median direction pixel value, refers to pixel to eight directions at semantic silhouette target edge, Mei Gefang
The intermediate value of upward all pixels value, is represented with MI.Similarly, eight directions one have eight MI values, and MI is designated as respectively1~MI8；Meter
Calculate formula as follows：
MIi=med { P1, P2..., Pn}
Wherein PjIt is pixel j spectral value, each direction has n pixel, j=1,2 ..., n.Med { } is intermediate value letter
Number.
In one embodiment of the invention, the S103 is based on vector marginal information and spatial domain scalogram picture is calculated and obtained
Codomain scalogram picture, including codomain scale-value, average-codomain two time scales approach calculated value are calculated by average-codomain two time scales approach
Domain scale-value is as follows：
For each pixel in remote sensing image, calculate on eight directions from the pixel to corresponding edge, each
The spectral value sum average value of all pixels point on direction, obtains ME1~ME8；
Calculate ME1~ME8Average value as the pixel codomain scale-value.
I.e. the pixel value of corresponding pixel points in codomain scalogram picture of the pixel is Ve；
Fig. 7 a are to calculate obtained image by average-codomain two time scales approach；
In one embodiment of the invention, the S103 is based on vector marginal information and spatial domain dimension calculation codomain chi
Degree, including codomain scale-value is calculated by intermediate value-codomain two time scales approach, this method is as described below：
For each pixel in the scalogram picture of spatial domain, eight directions from the pixel to corresponding edge are calculated
On, the intermediate value of the spectral value of all pixels point on each direction obtains MI1~MI8；
Calculate MI1~MI8Intermediate value as the pixel codomain scale-value Vm, i.e.,：
Vm=med { MI1,MIi,...,MI8}。
Fig. 7 b are to calculate obtained image by intermediate value-codomain two time scales approach.
In one embodiment of the invention, the S103 is based on vector marginal information and spatial domain dimension calculation codomain chi
Degree, including codomain scale-value is calculated by joint-codomain two time scales approach, this method is as follows：
For each pixel in the scalogram picture of spatial domain, calculate on eight directions from the pixel to corresponding edge,
The spectral value sum average value of all pixels point, obtains ME on each direction1~ME8；
Calculate ME1~ME8Intermediate value as the codomain scale-value, the i.e. pixel of the pixel codomain scale-value Vem=med
{ME1,ME2..., ME8}。
V hereinem、Ve、VmAll referring to codomain scale-value, lower footnote only represents the difference of calculating process, is not used to refer to
Show that the implication of their value is different.
For example, Fig. 7 c are to calculate obtained image by joint-codomain two time scales approach.
Present invention additionally comprises a kind of computer-readable recording medium, computer program is stored thereon with, the program is processed
The step of device realizes the above method when performing.
Present invention additionally comprises memory, processor and it is stored on the memory and can performs on the processor
Computer program, above-mentioned step is realized during the computing device described program
" at least one " used herein, " one or more " and "and/or" are open statements, when in use
It can be united and separation.For example, " in A, B and C at least one ", " in A, B or C at least one ", " in A, B and C
It is one or more " and " one or more of A, B or C " refer to only A, only B, only C, A and B together, A and C together,
B and C together or A, B and C together.
Term " one " entity refers to one or more entities.Thus term " one ", " one or more " and " extremely
Few one " be herein defined as can be with used interchangeably.It should also be noted that term " comprising ", "comprising" and " having " are also can be mutual
Change what is used.
Term " computer-readable medium " used herein refers to participate in providing instructions to any of computing device
Tangible storage device and/or transmission medium.During computer-readable medium can be network transmission (such as SOAP) on ip networks
The serial command collection of coding.Such medium can take many forms, and including but not limited to non-volatile media, volatibility is situated between
Matter and transmission medium.Non-volatile media includes such as NVRAM or magnetically or optically disk.Volatile media includes such as main storage
Dynamic memory (such as RAM).The common form of computer-readable medium includes such as floppy disk, flexible disk, hard disk, tape or appointed
What its magnetizing mediums, magnet-optical medium, CD-ROM, any other optical medium, punched card, paper tape, it is any other have hole shape pattern
Physical medium, RAM, PROM, EPROM, FLASH-EPROM, the solid state medium of such as storage card, any other storage chip or
Any other medium that cassette, the carrier wave described below or computer can be read.The digital file attachment of Email or
Other self-contained news files or archive set are considered as the distribution medium equivalent to tangible media.Work as computer-readable medium
When being configured as database, it should be appreciated that the database can be any kind of database, such as relational database, number of levels
According to storehouse, OODB Object Oriented Data Base etc..Correspondingly, it is believed that the present invention includes tangible media or distribution medium and existing skill
Equivalent known to art and the medium of following exploitation, store the software implementation of the present invention in these media.
Term used herein " it is determined that ", " computing " and " calculating " and its modification can be with used interchangeably, and including appointing
Method, processing, mathematical operation or the technology of what type.More specifically, the explanation that such term can include such as BPEL is advised
Then or rule language, wherein logic be not hard coded but in the rule file that can be read, explained, compiled and performed table
Show.
Term " module " used herein or " instrument " refer to hardware that is any of or developing later, software, consolidated
Part, artificial intelligence, fuzzy logic or be able to carry out the function related to the element hardware and software combination.In addition, though
The present invention is described with illustrative embodiments, it is to be understood that each aspect of the present invention can individually be claimed.
Term " comprising ", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that bag
Including process, method, article or the terminal device of a series of key elements not only includes those key elements, but also including not arranging clearly
Other key elements gone out, or also include for this process, method, article or the intrinsic key element of terminal device.Do not having
In the case of more limitations, the key element limited by sentence " including ... " or " including ... ", it is not excluded that including it is described will
Also there is other key element in process, method, article or the terminal device of element.
Although the various embodiments described above are described, those skilled in the art once know basic wound
The property made concept, then can make other change and modification to these embodiments, so embodiments of the invention are the foregoing is only,
Not thereby the scope of patent protection of the present invention, the equivalent structure that every utilization description of the invention and accompanying drawing content are made are limited
Or equivalent flow conversion, or other related technical fields are directly or indirectly used in, similarly it is included in the patent of the present invention
Within protection domain.
Claims (9)
1. a kind of remote sensing image dimension calculation method, it is characterised in that：Including：
S101 obtains vector marginal information；
S102 is based on vector marginal information and calculates spatial domain yardstick, obtains spatial domain scalogram picture；
S103 is based on vector marginal information and spatial domain scalogram picture calculates and obtains codomain scalogram picture.
2. according to the method described in claim 1, it is characterised in that include before step S101：
S201 is receiving remote sensing image；
S202 is pre-processed to remote sensing image, and pretreatment includes panchromatic image fusion treatment and multispectral image filtering process；
S203, to pretreated remote sensing image extract vector marginal information.
3. according to the method described in claim 1, it is characterised in that the S102 is based on vector marginal information and calculates spatial domain chi
Degree, obtaining spatial domain scalogram picture includes：
The distance that each pixel in remote sensing image arrives respective edges on K direction is calculated, pixel is arrived on K direction
The distance of respective edges is respectively N1~Nk；K is equal to 2 n powers, and n is more than or equal to 2；
The pixel value S of the corresponding pixel of the pixel in the spatial domain yardstick image of settingdFor N1~NkAverage value.
4. according to the method described in claim 1, it is characterised in that the S103 is based on vector marginal information and spatial domain scalogram
Codomain scalogram picture is obtained as calculating, including：For each pixel in remote sensing image, calculate from the pixel to corresponding sides
On K direction of edge, the spectral value sum average value of all pixels point on each direction obtains ME1~MEk；
Calculate ME1~MEkAverage value as the pixel codomain scale-value.
5. according to the method described in claim 1, it is characterised in that the S103 is based on vector marginal information and spatial domain yardstick meter
Codomain yardstick is calculated, including：
For each pixel in the scalogram picture of spatial domain, calculate from the pixel to the K direction at corresponding edge, each
The intermediate value of the spectral value of all pixels point on direction, obtains MI1~MIk；
Calculate MI1~MIkIntermediate value as the pixel codomain scale-value.
6. according to the method described in claim 1, it is characterised in that the S103 is based on vector marginal information and spatial domain yardstick meter
Codomain yardstick is calculated, including：
For each pixel in the scalogram picture of spatial domain, calculate on the K direction from the pixel to corresponding edge, Mei Gefang
The spectral value sum average value of upward all pixels point, obtains ME1~MEk；
Calculate ME1~MEkIntermediate value as the pixel codomain scale-value.
7. according to the method described in claim 1, it is characterised in that the multispectral image filtering process include medium filtering,
One or more in gaussian filtering, average drifting, bilateral filtering or anisotropic diffusion filtering.
8. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is held by processor
The step of methods described as any such as claim 1 to 7 is realized during row.
9. a kind of computer equipment, including memory, processor and it is stored on the memory and can be in the processor
The computer program of upper execution, it is characterised in that realized during the computing device described program as claim 1 to 7 is any
The step of methods described.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710301493.6A CN107274421B (en) | 2017-05-02 | 2017-05-02 | A kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710301493.6A CN107274421B (en) | 2017-05-02 | 2017-05-02 | A kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107274421A true CN107274421A (en) | 2017-10-20 |
| CN107274421B CN107274421B (en) | 2019-11-29 |
Family
ID=60073654
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710301493.6A Active CN107274421B (en) | 2017-05-02 | 2017-05-02 | A kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107274421B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583345A (en) * | 2018-11-21 | 2019-04-05 | 平安科技（深圳）有限公司 | Roads recognition method, device, computer installation and computer readable storage medium |
| CN112836467A (en) * | 2020-12-30 | 2021-05-25 | 腾讯科技（深圳）有限公司 | Image processing method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104200458A (en) * | 2014-07-30 | 2014-12-10 | 浙江工业大学 | MeanShift based high-resolution remote sensing image segmentation distance measurement optimization method |
| CN105528793A (en) * | 2016-01-12 | 2016-04-27 | 北京建筑大学 | High-spatial-resolution remote sensing image segmentation method and system |
-
2017
- 2017-05-02 CN CN201710301493.6A patent/CN107274421B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104200458A (en) * | 2014-07-30 | 2014-12-10 | 浙江工业大学 | MeanShift based high-resolution remote sensing image segmentation distance measurement optimization method |
| CN105528793A (en) * | 2016-01-12 | 2016-04-27 | 北京建筑大学 | High-spatial-resolution remote sensing image segmentation method and system |
Non-Patent Citations (2)
| Title |
|---|
| 刘建华: "多色彩高空间分辨率遥感影像矢量边缘信息提取算法与应用", 《地球信息科学学报》 * |
| 刘永学 等: "基于边缘的多光谱遥感图像分割方法", 《遥感学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583345A (en) * | 2018-11-21 | 2019-04-05 | 平安科技（深圳）有限公司 | Roads recognition method, device, computer installation and computer readable storage medium |
| CN109583345B (en) * | 2018-11-21 | 2023-09-26 | 平安科技（深圳）有限公司 | Road recognition method, device, computer device and computer readable storage medium |
| CN112836467A (en) * | 2020-12-30 | 2021-05-25 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN112836467B (en) * | 2020-12-30 | 2023-12-12 | 腾讯科技（深圳）有限公司 | Image processing method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107274421B (en) | 2019-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | An efficient approach for automatic rectangular building extraction from very high resolution optical satellite imagery | |
| Li et al. | An edge embedded marker-based watershed algorithm for high spatial resolution remote sensing image segmentation | |
| Isikdogan et al. | Automatic channel network extraction from remotely sensed images by singularity analysis | |
| Bi et al. | SAR image change detection using regularized dictionary learning and fuzzy clustering | |
| Aytekın et al. | Unsupervised building detection in complex urban environments from multispectral satellite imagery | |
| CN110781756A (en) | Urban road extraction method and device based on remote sensing image | |
| Abdollahi et al. | Extraction of road features from UAV images using a novel level set segmentation approach | |
| Ghorai et al. | Extracting shoreline from satellite imagery for GIS analysis | |
| Rishikeshan et al. | A novel mathematical morphology based algorithm for shoreline extraction from satellite images | |
| Chen et al. | A new process for the segmentation of high resolution remote sensing imagery | |
| CN107705313A (en) | A kind of remote sensing images Ship Target dividing method | |
| CN107451595A (en) | Infrared image salient region detection method based on hybrid algorithm | |
| CN104299241A (en) | Remote sensing image significance target detection method and system based on Hadoop | |
| Zhang et al. | Level set evolution with local uncertainty constraints for unsupervised change detection | |
| CN107274421B (en) | A kind of remote sensing image dimension calculation method, readable storage medium storing program for executing and computer equipment | |
| CN103236052A (en) | Automatic cell localization method based on minimized model L1 | |
| Wang et al. | A region-line primitive association framework for object-based remote sensing image analysis | |
| Yu et al. | MSFTrans: A multi-task frequency-spatial learning transformer for building extraction from high spatial resolution remote sensing images | |
| Lê et al. | Learning digital terrain models from point clouds: ALS2DTM dataset and rasterization-based gan | |
| Zhao et al. | Image network teaching resource retrieval algorithm based on deep hash algorithm | |
| CN112150474A (en) | Underwater bubble image feature segmentation and extraction method | |
| CN116012393A (en) | Carton point cloud segmentation method, device and processing equipment | |
| Liu et al. | Adaptive algorithm for automated polygonal approximation of high spatial resolution remote sensing imagery segmentation contours | |
| Xumin et al. | Research on vectorization of weather radar image | |
| Wang et al. | RSBuilding: Towards General Remote Sensing Image Building Extraction and Change Detection with Foundation Model |
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 |