CN108376232A - A kind of method and apparatus of automatic interpretation for remote sensing image - Google Patents
A kind of method and apparatus of automatic interpretation for remote sensing image Download PDFInfo
- Publication number
- CN108376232A CN108376232A CN201810007957.7A CN201810007957A CN108376232A CN 108376232 A CN108376232 A CN 108376232A CN 201810007957 A CN201810007957 A CN 201810007957A CN 108376232 A CN108376232 A CN 108376232A
- Authority
- CN
- China
- Prior art keywords
- block message
- remote sensing
- surface type
- ground surface
- ground block
- 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.)
- Withdrawn
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
- G06V20/13—Satellite images
-
- 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
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Abstract
The object of the present invention is to provide a kind of method and apparatus of the automatic interpretation for remote sensing image.The present invention is by obtaining the ground block message in remote sensing image, then the Raster Images data corresponding to described ground block message are obtained, the characteristics of image in the Raster Images data is extracted, finally according to described image feature, determines the ground surface type corresponding to described ground block message.Compared with prior art, the present invention realizes the automatic interpretation for remote sensing image, reduces calculation amount and improves the accuracy of interpretation.
Description
Technical field
The present invention relates to technical field of remote sensing image processing more particularly to a kind of skills of automatic interpretation for remote sensing image
Art.
Background technology
Remote sensing image (Remote Sensing Image) refers to the film or photograph for recording various atural object electromagnetic wave sizes,
Such as airphoto and satellite photograph.All include corresponding geographical coordinate in every remote sensing image, it is in other words, every in remote sensing image
A pixel can obtain its unique geographical coordinate position.
Since remote sensing image can reflect various atural object images, how to obtain the terrestrial object information in remote sensing image
Become a problem.In the prior art, the terrestrial object information in remote sensing image is obtained mainly by way of human interpretation,
I.e.:First, ground block message is obtained by field survey and obtains the remote sensing image of covering ground block message, then, for remote sensing shadow
As carrying out human interpretation, atural object familygram is obtained, finally, by atural object familygram with ground block message combination processing, obtains plot letter
Interpretation result inside breath.
However defect certain existing for this problem:First, " field survey block message " and " human interpretation " etc.
Process is required to a large amount of artificial mark work;Second, exist for remote sensing image and computes repeatedly;Third, " by ground species
Figure same ground block message combination processing " content is cumbersome, computationally intensive.
Therefore, hand labor how is effectively reduced or avoided, quickly and accurately realizes the automatic neutralizing of remote sensing image
It translates, becomes those skilled in the art institute urgent problem to be solved.
Invention content
The object of the present invention is to provide a kind of method and apparatus of the automatic interpretation for remote sensing image.
According to one embodiment of present invention, a kind of method of the automatic interpretation for remote sensing image is provided, wherein should
Method includes the following steps:
Obtain one or more of remote sensing image ground block message;
Obtain the Raster Images data corresponding to described ground block message;
Extract the characteristics of image in the Raster Images data;
According to described image feature, the ground surface type corresponding to described ground block message is determined.
Optionally, the step of one or more of acquisition remote sensing image ground block message includes:
Image preprocessing is carried out to remote sensing images, to generate remote sensing images to be detected;
Straight-line detection is carried out to the remote sensing image to be detected, with extract one on the remote sensing images to be detected or
A plurality of straight line;
Linear optimal processing is carried out to the straight line, optimizes straight line to determine;
According to the optimization straight line, the profile in the remote sensing images is extracted;
The profile extracted is fitted, to generate one or more ground block message in the remote sensing images.
Optionally, the step of carrying out image preprocessing to remote sensing images includes following at least any one:
Image gray processing is carried out to the remote sensing images;
Bilateral filtering is carried out to the remote sensing images;
Histogram equalization is carried out to the remote sensing images;
Image sharpening is carried out to the remote sensing images.
Optionally, include to the step of remote sensing image progress straight-line detection to be detected:
Based on LSD algorithm, straight-line detection is carried out to the remote sensing image to be detected, to extract the remote sensing figure to be detected
As one or more upper straight line.
Optionally, include to the step of straight line progress linear optimal processing:
The coordinate of every straight line is converted into polar coordinates;
According to the polar coordinates of the straight line, the straight line is clustered, to generate one or more straight line clusters;
Determine extended line corresponding with the straight line cluster;
According to predetermined threshold, the extended line is filtered, to obtain one or more optimization straight lines.
Optionally, the step of extracting the profile in the remote sensing images include:
By the mask of the optimization straight line gernertion binaryzation;
In the mask, the profile in the remote sensing images is extracted.
Optionally, the step of profile extracted being fitted include:
The profile extracted is filtered out;
Whole profiles after being filtered out execute convex closure fitting, to generate one or more plot in the remote sensing images
Information.
Optionally, obtain described ground block message corresponding to Raster Images data the step of include:
According to described ground block message, operation is split to the remote sensing image corresponding to described ground block message, to obtain
State the Raster Images data corresponding to ground block message.
Optionally, the step of extracting the characteristics of image in the Raster Images data include:
According to the original band class information corresponding to the Raster Images data, the image in the Raster Images data is extracted
Feature.
Optionally, described image feature includes following at least any one:
The derivation information of the original band class information of the Raster Images data and the original band class information;
The index information of the Raster Images data;
The texture information of the Raster Images data.
Optionally it is determined that the step of ground surface type corresponding to described ground block message, includes:
According to described image feature, the candidate ground surface type of one or more corresponding to described ground block message is determined;
According to the candidate ground surface type, the earth's surface in the one or more associations plot in conjunction with corresponding to described ground block message
Type information determines the ground surface type corresponding to described ground block message.
Optionally it is determined that the step of one or more corresponding to described ground block message candidate ground surface type, includes:
According to described image feature, in conjunction with scheduled disaggregated model, one or more corresponding to described ground block message is determined
A candidate's ground surface type, wherein candidate's ground surface type includes candidate ground surface type title and corresponding candidate earth's surface class
Type probability.
Optionally, in conjunction with the position related information between multiple plot with the earth's surface class corresponding to determination described ground block message
The step of type includes:
According to the position incidence relation between described ground block message and other ground block messages, determine that described ground block message institute is right
The one or more associations plot answered;
According to the candidate ground surface type described ground block message is determined in conjunction with the ground surface type information in the association plot
Corresponding ground surface type.
According to another embodiment of the invention, a kind of interpretation of the automatic interpretation for remote sensing image is additionally provided to set
It is standby, wherein the interpretation device includes:
First device, for obtaining one or more of remote sensing image ground block message;
Second device, for obtaining the Raster Images data corresponding to described ground block message;
3rd device, for extracting the characteristics of image in the Raster Images data;
4th device, for according to described image feature, determining the ground surface type corresponding to described ground block message.
Optionally, the first device includes:
Unit one by one, for carrying out image preprocessing to remote sensing images, to generate remote sensing images to be detected;
Unit one or two, for carrying out straight-line detection to the remote sensing image to be detected, to extract the remote sensing to be detected
One or more straight line on image;
Unit one or three optimize straight line for carrying out linear optimal processing to the straight line to determine;
Unit one or four, for according to the optimization straight line, extracting the profile in the remote sensing images;
First Five-Year Plan unit, for the profile extracted to be fitted, to generate one or more in the remote sensing images
Ground block message.
Optionally, the unit one by one is for following at least any one:
Image gray processing is carried out to the remote sensing images;
Bilateral filtering is carried out to the remote sensing images;
Histogram equalization is carried out to the remote sensing images;
Image sharpening is carried out to the remote sensing images.
Optionally, Unit one or two is used for:
Based on LSD algorithm, straight-line detection is carried out to the remote sensing image to be detected, to extract the remote sensing figure to be detected
As one or more upper straight line.
Optionally, Unit one or three is used for:
The coordinate of every straight line is converted into polar coordinates;
According to the polar coordinates of the straight line, the straight line is clustered, to generate one or more straight line clusters;
Determine extended line corresponding with the straight line cluster;
According to predetermined threshold, the extended line is filtered, to obtain one or more optimization straight lines.
Optionally, Unit one or four is used for:
By the mask of the optimization straight line gernertion binaryzation;
In the mask, the profile in the remote sensing images is extracted.
Optionally, the First Five-Year Plan unit is used for:
The profile extracted is filtered out;
Whole profiles after being filtered out execute convex closure fitting, to generate one or more plot in the remote sensing images
Information.
Optionally, the second device is used for:
According to described ground block message, operation is split to the remote sensing image corresponding to described ground block message, to obtain
State the Raster Images data corresponding to ground block message.
Optionally, the 3rd device is used for:
According to the original band class information corresponding to the Raster Images data, the image in the Raster Images data is extracted
Feature.
Optionally, described image feature includes following at least any one:
The derivation information of the original band class information of the Raster Images data and the original band class information;
The index information of the Raster Images data;
The texture information of the Raster Images data.
Optionally, the 4th device includes:
Unit 41, for according to described image feature, determining that the one or more corresponding to described ground block message is candidate
Ground surface type;
Unit four or two are used for according to the candidate ground surface type, the one or more in conjunction with corresponding to described ground block message
It is associated with the ground surface type information in plot, determines the ground surface type corresponding to described ground block message.
Optionally, Unit 41 is used for:
According to described image feature, in conjunction with scheduled disaggregated model, one or more corresponding to described ground block message is determined
A candidate's ground surface type, wherein candidate's ground surface type includes candidate ground surface type title and corresponding candidate earth's surface class
Type probability.
Optionally, Unit four or two is used for:
According to the position incidence relation between described ground block message and other ground block messages, determine that described ground block message institute is right
The one or more associations plot answered;
According to the candidate ground surface type described ground block message is determined in conjunction with the ground surface type information in the association plot
Corresponding ground surface type.
Compared with prior art, then the present invention obtains the plot letter by obtaining the ground block message in remote sensing image
The corresponding Raster Images data of breath, extract the characteristics of image in the Raster Images data, finally according to described image feature,
Determine the ground surface type corresponding to described ground block message;To which the present invention realizes the automatic interpretation for remote sensing image, reduce
Calculation amount and the accuracy for improving interpretation.
Description of the drawings
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, of the invention other
Feature, objects and advantages will become more apparent upon:
Fig. 1 shows to be illustrated according to a kind of interpretation device of automatic interpretation for remote sensing image of one aspect of the invention
Figure;
Fig. 2 shows a kind of interpretations of automatic interpretation for remote sensing image according to a preferred embodiment of the present invention to set
Standby schematic diagram;
Fig. 3 shows a kind of method flow diagram of automatic interpretation for remote sensing image according to a further aspect of the present invention;
Fig. 4 shows a kind of method stream of automatic interpretation for remote sensing image according to a preferred embodiment of the present invention
Cheng Tu.
Same or analogous reference numeral represents same or analogous component in attached drawing.
Specific implementation mode
It should be mentioned that some exemplary embodiments are described as before exemplary embodiment is discussed in greater detail
The processing described as flow chart or method.Although operations are described as the processing of sequence by flow chart, therein to be permitted
Multioperation can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of operations can be rearranged.When it
The processing can be terminated when operation completion, it is also possible to the additional step being not included in attached drawing.The processing
It can correspond to method, function, regulation, subroutine, subprogram etc..
So-called within a context " interpretation device ", as " computer equipment ", also referred to as " computer " refer to that can pass through
Operation preset program instructs to execute the intelligent electronic device of the predetermined process process such as numerical computations and/or logical calculated,
It may include processor and memory, the survival to prestore in memory executed by processor and is instructed to execute predetermined process mistake
Journey, or predetermined process process is executed by hardware such as ASIC, FPGA, DSP, or realized by said two devices combination.
The computer equipment includes user equipment and/or the network equipment.Wherein, the user equipment includes but not limited to
Computer, smart mobile phone, PDA etc.;The network equipment includes but not limited to single network server, multiple network servers composition
Server group or based on cloud computing (Cloud Computing) the cloud being made of a large amount of computers or network server,
In, cloud computing is one kind of Distributed Calculation, a super virtual computer being made of the computer collection of a group loose couplings.
Wherein, the computer equipment can isolated operation realize the present invention, also can access network and by with other meters in network
The interactive operation of machine equipment is calculated to realize the present invention.Wherein, the network residing for the computer equipment includes but not limited to interconnect
Net, wide area network, Metropolitan Area Network (MAN), LAN, VPN network etc..
Those skilled in the art will be understood that under normal circumstances heretofore described " interpretation device " can only be net
Network equipment is executed corresponding operation by the network equipment;Under special circumstances, can also be by user equipment and the network equipment
Or server is integrated to form, i.e., is matched to execute corresponding operation, for example, by user with the network equipment by user equipment
Equipment sends to the network equipment and instructs, to indicate that the network equipment starts the corresponding operating of execution " automatic interpretation of remote sensing image ".
It should be noted that the user equipment, the network equipment and network etc. are only for example, other are existing or from now on may be used
The computer equipment or network that can occur such as are applicable to the present invention, should also be included within the scope of the present invention, and to draw
It is incorporated herein with mode.
Specific structure and function details disclosed herein are only representative, and are for describing the present invention show
The purpose of example property embodiment.But the present invention can be implemented by many alternative forms, and be not interpreted as
It is limited only by the embodiments set forth herein.
Although it should be understood that may have been used term " first ", " second " etc. herein to describe each unit,
But these units should not be limited by these terms.The use of these items is only for by a unit and another unit
It distinguishes.For example, without departing substantially from the range of exemplary embodiment, it is single that first unit can be referred to as second
Member, and similarly second unit can be referred to as first unit.Term "and/or" used herein above include one of them or
The arbitrary and all combination of more listed associated items.
Term used herein above is not intended to limit exemplary embodiment just for the sake of description specific embodiment.Unless
Context clearly refers else, otherwise singulative used herein above "one", " one " also attempt to include plural number.Also answer
When understanding, term " include " and or " include " used herein above provide stated feature, integer, step, operation,
The presence of unit and/or component, and do not preclude the presence or addition of other one or more features, integer, step, operation, unit,
Component and/or a combination thereof.
It should further be mentioned that in some replace implementations, the function action being previously mentioned can be according to different from attached
The sequence indicated in figure occurs.For example, involved function action is depended on, the two width figures shown in succession actually may be used
Substantially simultaneously to execute or can execute in a reverse order sometimes.
Present invention is further described in detail below in conjunction with the accompanying drawings.
Fig. 1 shows to be illustrated according to a kind of interpretation device of automatic interpretation for remote sensing image of one aspect of the invention
Figure;Wherein, the interpretation device includes first device 1, second device 2,3rd device 3, the 4th device 4.
Specifically, the first device 1 obtains one or more of remote sensing image ground block message;The second device 2
Obtain the Raster Images data corresponding to described ground block message;The 3rd device 3 extracts the figure in the Raster Images data
As feature;4th device 4 determines the ground surface type corresponding to described ground block message according to described image feature.
The first device 1 obtains one or more of remote sensing image ground block message.
Specifically, the first device 1 by way of marking on the spot, to obtain one or more of remote sensing image ground
Block message;Wherein, the mark on the spot refers to, in region to be measured, being described by hand-held instrument corresponding by operator
The contour edge of atural object, to obtain in remote sensing image block message method.
Alternatively, the first device 1 is by one or more pieces of partitioning algorithms, by carrying out plot segmentation to remote sensing images
After obtain one or more of remote sensing image ground block message.Wherein, described piece of partitioning algorithm includes but not limited to be based on side
Plot extracting method, the plot extracting method based on segmentation and the plot extracting method based on lines detection of edge detection, on
It includes but not limited to image gray processing to state in method, image binaryzation, Edge extraction, image outline fitting, image, semantic
The process contents such as segmentation.
The form of the ground block message obtained by using mark on the spot or block partitioning algorithm is generally the set of multigroup point,
Every group of point represents the minimum outer boundary in a block region, put geographical location information included in information include but are not limited to through
Latitude information, projected coordinate system information etc..The set of multigroup point constitutes whole ground block message.
Those skilled in the art will be understood that plot information of the present invention includes but not limited to farmland massif, city
The arbitrarily block message such as plot.
Preferably, the first device 1 can carry out image preprocessing to remote sensing images, to generate remote sensing figure to be detected
Picture;Straight-line detection is carried out to the remote sensing image to be detected, to extract one or more on the remote sensing images to be detected
Straight line;Linear optimal processing is carried out to the straight line, optimizes straight line to determine;According to the optimization straight line, the remote sensing is extracted
Profile in image;The profile extracted is fitted, to generate one or more ground block message in the remote sensing images.
Specifically, after the first device 1 obtains pending remote sensing images, such as pass through filtering, digitlization, geometry change
Change, normalize, smoothly, restore, enhancing etc. image procossings mode, to the remote sensing images carry out image preprocessing, with generate wait for
Detect remote sensing images.
Preferably, the pretreated mode of described image includes but not limited to following at least any one:
Image gray processing is carried out to the remote sensing images;
Bilateral filtering is carried out to the remote sensing images;
Histogram equalization is carried out to the remote sensing images;
Image sharpening is carried out to the remote sensing images.
It is highly preferred that image preprocessing can be carried out to remote sensing images according to processing sequence below:
First acquired pending remote sensing images gray processing;Then, the remote sensing images after gray processing are carried out double
Side filters, and edge is still maintained after obtaining the image similar to Gaussian Blur;Next, to bilateral filtering treated remote sensing
Image carries out histogram equalization, image color is balanced, in favor of subsequent detection;Finally, using Laplace operator, to distant
Sense image is filtered, and is equivalent to and is enhanced edge, has carried out image sharpening.
Since linear character is typically all edge feature, it is more conducive to it by above-mentioned pretreated remote sensing images
Straight-line detection afterwards.
Then, after image preprocessing, the first device 1 carries out straight-line detection to the remote sensing image to be detected, with
Extract one or more straight line on the remote sensing images to be detected.Here, a variety of line detection algorithms can be utilized, such as
LSD line detection algorithms, Freeman straight-line detections, cankerworm-crawl algorithm etc. carry out pretreated remote sensing image to be detected
Straight-line detection, to extract one or more straight line on the remote sensing images to be detected.
After extracting straight line, the first device 1 by the way that the multiple straight lines or linear segments that are generated are clustered, and
Straight line after cluster arrange and denoising optimizes straight line to realize the optimization processing to the straight line to determine.
Preferably, the coordinate of every straight line can be converted to polar coordinates by the first device 1;According to the straight line
Polar coordinates, the straight line is clustered, to generate one or more straight line clusters;Determine prolong corresponding with the straight line cluster
Long line;According to length predetermined threshold, the extended line is filtered, to obtain one or more optimization straight lines.
Specifically, the coordinate corresponding to every straight line is transformed into the polar coordinates under polar coordinate system by the first device 1, often
Straight line is a point under polar coordinate system;Then, according to r in the polar coordinates of the straight line (i.e. the coordinate of polar coordinates point)
The value of (radial coordinate) and theta (polar angle) are clustered, to which the smaller straight line of gap is clustered into a straight line cluster;To every
Cluster straight line calculates the mean value of its r (radial coordinate) and theta (polar angle), and determines the beginning and end of the cluster straight line, according to
Every straight line at a distance from straight line corresponding to mean value into the judgement for being about to every straight line and receiving or filtering out, so that it is determined that with described straight
The corresponding extended line of line cluster;Finally, according to default setting or the length predetermined threshold based on determined by straight line cluster, prolong to described
Long line is filtered, and even extended line is less than the length predetermined threshold, that is, is filtered out, to delete straight line fine crushing, to obtain
Take one or more optimization straight lines.
After determining optimization straight line, the first device 1 is based on all kinds of contours extract algorithms according to the optimization straight line,
Extract the profile in the remote sensing images.Preferably, the optimization straight line is widened or is slightly added by the first device 1
Width, to generate the mask (MASK) of binaryzation;Then in the mask generated, the profile in the remote sensing images is extracted.
Finally, all profiles extracted are carried out convex closure fitting by the first device 1, after fitting result is exported, i.e.,
Generate one or more ground block message in the remote sensing images.
Preferably, the first device 1 can filter out the profile extracted, for example, by profile it is smaller before hundred
The little profile of/N is deleted, alternatively, by profile before smaller 10% little profile it is similar according to histogram to neighbour's profile
Degree merges, to achieve the effect that filter out.Then, whole profiles after filtering out carry out convex closure fitting, and fitting result is defeated
After going out, that is, generate one or more ground block message in the remote sensing images.
The second device 2 obtains the Raster Images data corresponding to described ground block message.
Specifically, the second device 2 is for the image data on specified ground block message by carrying out rasterizing, then into
Row trimming operation with the Raster Images data on extraction ground block message, and is stored with specified data format.Wherein, institute
State the ground residing for the band class information, geographical location information, resolution information, grid that Raster Images data include but not limited to grid
Block message etc..
Here, the mode for obtaining Raster Images data includes but not limited to Grid Method, transformation approach, scanning digital method or classification
The methods of image input.
Preferably, the second device 2 can be according to described ground block message, to the remote sensing shadow corresponding to described ground block message
As being split operation, to obtain the Raster Images data corresponding to described ground block message.
Specifically, the second device 2 is according to different ground block messages, such as the position of block message, block message it is big
It is small, block message and remaining ground block message the information such as relevance, determine and the corresponding different rasterizings of differently block message
Segmentation attribute information, such as segmentation precision, the area size of segmentation, to obtain the grid corresponding to described ground block message
Image data.
For example, if the position of a certain ground block message corresponds to the region for needing high accuracy analysis, the precision of segmentation is improved;
If the area of a certain ground block message is more than another plot, the segmentation precision corresponding to the plot of small area is improved;If a certainly
The relevance of block message and remaining ground block message is very high, then the segmentation precision of the ground block message can be improved, or be based on its leeway
Segmentation precision corresponding to block message, to determine the segmentation precision etc. in current plot.
It preferably, can be in conjunction with the geographical location corresponding to described ground block message when obtaining the Raster Images data
Information determines the information such as the segmentation precision of extraction Raster Images data, thus based on identified processing mode, it is described to obtain
Raster Images data corresponding to ground block message.
The 3rd device 3 extracts the characteristics of image in the Raster Images data.
Specifically, the 3rd device 3 carries out feature extraction for each acquired Raster Images data, to obtain
Characteristics of image in the Raster Images data.Those skilled in the art will be understood that when extracting different characteristics of image, institute
Corresponding extracting method is different.
Preferably, the 3rd device 3 can be extracted according to the original band class information corresponding to the Raster Images data
Characteristics of image in the Raster Images data.
Specifically, the 3rd device 3 can be processed the original band class information of the Raster Images data, and from
Described image feature is extracted in original band class information.
It is highly preferred that described image feature includes following at least any one:
The derivation information of the original band class information of the Raster Images data and the original band class information:The derivative
Information include but not limited to the original band class information is normalized, binaryzation or color space conversion etc. it is one or more
After processing, the information that is obtained.
The index information of the Raster Images data:Calculated using the original band combination of the Raster Images data
The index information gone out.The index information includes but not limited to:NDMI, i.e. difference humidity index (vegetation water content);NDVI, i.e.,
Normalized differential vegetation index;ARVI, i.e. air impedance vegetation index;NBR, i.e. normalization burning ratio;NBR2, i.e. normalization burning
Ratio 2;EVI, i.e. environmental vegetation index;SR, i.e. ratio vegetation index etc..
The texture information of the Raster Images data:The texture information mainly include Raster Images data adjacent picture elements it
Between relationship, i.e., feature the pixel extracted by modes such as convolution, ponds between Raster Images data.
4th device 4 determines the ground surface type corresponding to described ground block message according to described image feature.
Specifically, the 4th device 4 utilizes intelligent algorithm, the model that described image characteristic use is trained in advance
Classify, by the way that described image feature to be compared with ground surface type, with the earth's surface corresponding to determination described ground block message
Type.
Fig. 2 shows a kind of interpretations of automatic interpretation for remote sensing image according to a preferred embodiment of the present invention to set
Standby schematic diagram;Wherein, the interpretation device includes first device 1, second device 2,3rd device 3, the 4th device 4, and described
Four devices 4 include 41 units 41 and four or two units 42.
Specifically, the first device 1 obtains one or more of remote sensing image ground block message;The second device 2
Obtain the Raster Images data corresponding to described ground block message;The 3rd device 3 extracts the figure in the Raster Images data
As feature;Unit 41 41 determines that the one or more corresponding to described ground block message is candidate according to described image feature
Ground surface type;Unit four or two 42 is according to the candidate ground surface type, one or more in conjunction with corresponding to described ground block message
The ground surface type information in a association plot, determines the ground surface type corresponding to described ground block message.
Wherein, the first device 1, the second device 2 and the 3rd device 3 and the corresponding intrument described in Fig. 1
It is same or similar, therefore details are not described herein, and be incorporated herein by reference.
Unit 41 41 determines that the one or more corresponding to described ground block message is candidate according to described image feature
Ground surface type.
Specifically, Unit 41 41 utilizes intelligent algorithm, the mould that described image characteristic use is trained in advance
Type is classified, by the way that described image feature to be compared with ground surface type, with one corresponding to determination described ground block message
A or multiple candidate ground surface type ground surface types.
Preferably, Unit 41 41 can according to described image feature, in conjunction with scheduled disaggregated model, determine described in
The candidate ground surface type of one or more corresponding to ground block message, wherein candidate's ground surface type includes candidate ground surface type
Title and corresponding candidate ground surface type probability.
Specifically, the disaggregated model includes but not limited to vector machine, convolutional neural networks, random forest, linear classification
Device etc., in obtained candidate's ground surface type, including the title of candidate ground surface type and corresponding candidate ground surface type probability.
Here, according to a kind of disaggregated model, then the candidate ground surface type title and the candidate ground surface type probability are directly corresponding;
According to a variety of disaggregated models, then the probability corresponding to same candidate ground surface type title that a variety of disaggregated models can be calculated
Be weighted or weighted average after, determine the candidate ground surface type probability corresponding to candidate's ground surface type title.
Unit four or two 42 is according to the candidate ground surface type, the one or more in conjunction with corresponding to described ground block message
It is associated with the ground surface type information in plot, determines the ground surface type corresponding to described ground block message.
Earth's surface corresponding to one or more associations plot of the Unit four or two 42 corresponding to the block message of current position
Type information selects the plot in conjunction with the candidate ground surface type in the current plot from the candidate ground surface type
Ground surface type corresponding to information.Wherein, the association plot includes but not limited to that the adjacent plot in current plot, shape are similar
Plot, the similar plot of feature etc., the ground surface type information in the association plot has been identified or has obtained more
A candidate's ground surface type information, to, Unit four or two 42 according to the ground surface type information in the association plot come from current
Selection confirms ground surface type in the candidate ground surface type in plot, or is associated with described in the candidate ground surface type in plot, probability
Highest one as reference, with the ground surface type in the current plot of determination.
Here, the ground surface type in the association plot can be identical as the ground surface type of current position block message or not
Together, for example, the ground surface type in plot will can be directly associated with as the ground surface type of the current position block message, alternatively, for example,
Can be facing-port industry area by adjacent building Parcel division if the ground surface type for being associated with plot is ocean.
Preferably, Unit four or two 42 can be associated with according to the position between described ground block message and other ground block messages
Relationship determines the one or more associations plot corresponding to described ground block message;According to the candidate ground surface type, in conjunction with described
It is associated with the ground surface type information in plot, determines the ground surface type corresponding to described ground block message.
Specifically, Unit four or two 42 can be associated with according to the position between described ground block message and other ground block messages
Relationship, will be with plot of the position range of current position block message in certain threshold value as being associated with plot.Here, the position model
It encloses including but not limited at a distance from the current position block message in certain threshold value, or is such as in Same Latitude or same
On the relative positions such as longitude;Further, the reference factors such as administrative division are can be combined with, to determine association plot.
Then, Unit four or two 42 is by the rule of formulation, according to the candidate ground surface type, in conjunction with it is described associatedly
The ground surface type information of block, determines the ground surface type corresponding to described ground block message.If the ground surface type for being associated with plot is ocean,
Can be then facing-port industry area by adjacent building Parcel division.
Fig. 3 shows a kind of method flow diagram of automatic interpretation for remote sensing image according to a further aspect of the present invention.
Specifically, in step sl, the interpretation device obtains one or more of remote sensing image ground block message;In step
In rapid S2, the interpretation device obtains the Raster Images data corresponding to described ground block message;In step s3, the interpretation is set
The standby characteristics of image extracted in the Raster Images data;In step s 4, the interpretation device is according to described image feature, really
Ground surface type corresponding to fixed described ground block message.
In step sl, the interpretation device obtains one or more of remote sensing image ground block message.
Specifically, in step sl, the interpretation device by way of marking on the spot, to obtain one in remote sensing image
A or multiple ground block message;Wherein, the mark on the spot refers to, in region to be measured, passing through hand-held instrument by operator
The contour edge for describing corresponding atural object, to obtain in remote sensing image block message method.
Alternatively, in step sl, the interpretation device by one or more pieces of partitioning algorithms, by remote sensing images into
One or more of remote sensing image ground block message is obtained after the segmentation of row plot.Wherein, described piece of partitioning algorithm includes but not
It is limited to the plot extracting method based on edge detection, the plot extracting method based on segmentation and the plot based on lines detection to carry
It takes method, includes but not limited to image gray processing in the above method, image binaryzation, Edge extraction, image outline fitting,
The process contents such as image, semantic segmentation.
The form of the ground block message obtained by using mark on the spot or block partitioning algorithm is generally the set of multigroup point,
Every group of point represents the minimum outer boundary in a block region, put geographical location information included in information include but are not limited to through
Latitude information, projected coordinate system information etc..The set of multigroup point constitutes whole ground block message.
Those skilled in the art will be understood that plot information of the present invention includes but not limited to farmland massif, city
The arbitrarily block message such as plot.
Preferably, in step sl, the interpretation device can carry out remote sensing images image preprocessing, to be checked to generate
Survey remote sensing images;Straight-line detection is carried out to the remote sensing image to be detected, to extract one on the remote sensing images to be detected
Item or a plurality of straight line;Linear optimal processing is carried out to the straight line, optimizes straight line to determine;According to the optimization straight line, extraction
Profile in the remote sensing images;The profile extracted is fitted, to generate one or more in the remote sensing images
Ground block message.
Specifically, in step sl, after the interpretation device obtains pending remote sensing images, such as by filtering, counting
The image procossings mode such as word, geometric transformation, normalization, smooth, recovery, enhancing carries out image to the remote sensing images and locates in advance
Reason, to generate remote sensing images to be detected.
Preferably, the pretreated mode of described image includes but not limited to following at least any one:
Image gray processing is carried out to the remote sensing images;
Bilateral filtering is carried out to the remote sensing images;
Histogram equalization is carried out to the remote sensing images;
Image sharpening is carried out to the remote sensing images.
It is highly preferred that image preprocessing can be carried out to remote sensing images according to processing sequence below:
First acquired pending remote sensing images gray processing;Then, the remote sensing images after gray processing are carried out double
Side filters, and edge is still maintained after obtaining the image similar to Gaussian Blur;Next, to bilateral filtering treated remote sensing
Image carries out histogram equalization, image color is balanced, in favor of subsequent detection;Finally, using Laplace operator, to distant
Sense image is filtered, and is equivalent to and is enhanced edge, has carried out image sharpening.
Since linear character is typically all edge feature, it is more conducive to it by above-mentioned pretreated remote sensing images
Straight-line detection afterwards.
Then, after image preprocessing, the interpretation device carries out straight-line detection to the remote sensing image to be detected, to carry
Take out one or more straight line on the remote sensing images to be detected.Here, a variety of line detection algorithms, such as LSD can be utilized
Line detection algorithm, Freeman straight-line detections, cankerworm-crawl algorithm etc. carry out pretreated remote sensing image to be detected straight
Line detects, to extract one or more straight line on the remote sensing images to be detected.
After extracting straight line, the interpretation device by the way that the multiple straight lines or linear segments that are generated are clustered, and
Straight line after cluster arrange and denoising optimizes straight line to realize the optimization processing to the straight line to determine.
Preferably, in step sl, the coordinate of every straight line can be converted to polar coordinates by the interpretation device;Root
According to the polar coordinates of the straight line, the straight line is clustered, to generate one or more straight line clusters;It determines and the straight line cluster
Corresponding extended line;According to length predetermined threshold, the extended line is filtered, it is straight to obtain one or more optimizations
Line.
Specifically, the coordinate corresponding to every straight line is transformed into the polar coordinates under polar coordinate system by the interpretation device, often
Straight line is a point under polar coordinate system;Then, according to r in the polar coordinates of the straight line (i.e. the coordinate of polar coordinates point)
The value of (radial coordinate) and theta (polar angle) are clustered, to which the smaller straight line of gap is clustered into a straight line cluster;To every
Cluster straight line calculates the mean value of its r (radial coordinate) and theta (polar angle), and determines the beginning and end of the cluster straight line, according to
Every straight line at a distance from straight line corresponding to mean value into the judgement for being about to every straight line and receiving or filtering out, so that it is determined that with described straight
The corresponding extended line of line cluster;Finally, according to default setting or the length predetermined threshold based on determined by straight line cluster, prolong to described
Long line is filtered, and even extended line is less than the length predetermined threshold, that is, is filtered out, to delete straight line fine crushing, to obtain
Take one or more optimization straight lines.
After determining optimization straight line, the interpretation device is based on all kinds of contours extract algorithms, carries according to the optimization straight line
Take the profile in the remote sensing images.Preferably, the optimization straight line is widened or is slightly widened by the interpretation device, with
Generate the mask (MASK) of binaryzation;Then in the mask generated, the profile in the remote sensing images is extracted.
Finally, all profiles extracted are carried out convex closure fitting by the interpretation device, after fitting result is exported, i.e., raw
At one or more ground block message in the remote sensing images.
Preferably, in step sl, the interpretation device can filter out the profile extracted, for example, by profile
The little profile of smaller preceding percent N is deleted, alternatively, by profile before smaller 10% little profile and neighbour's profile according to
Histogram similarity merges, to achieve the effect that filter out.Then, whole profiles after filtering out carry out convex closure fitting, will
After fitting result output, that is, generate one or more ground block message in the remote sensing images.
In step s 2, the interpretation device obtains the Raster Images data corresponding to described ground block message.
Specifically, in step s 2, the interpretation device for the image data on specified ground block message by carrying out grid
It formats, then carries out trimming operation, with the Raster Images data on extraction ground block message, and deposited with specified data format
Storage.Wherein, the Raster Images data include but not limited to the band class information of grid, geographical location information, resolution information, grid
Ground block message residing for lattice etc..
Here, the mode for obtaining Raster Images data includes but not limited to Grid Method, transformation approach, scanning digital method or classification
The methods of image input.
Preferably, in step s 2, the interpretation device can be according to described ground block message, and to described ground block message, institute is right
The remote sensing image answered is split operation, to obtain the Raster Images data corresponding to described ground block message.
Specifically, in step s 2, the interpretation device is according to different ground block messages, as block message position,
The size of block message, block message and remaining ground block message the information such as relevance, determination is corresponding with differently block message
The segmentation attribute information of different rasterizings, such as the precision of segmentation, the area size of segmentation, to obtain described ground block message institute
Corresponding Raster Images data.
For example, if the position of a certain ground block message corresponds to the region for needing high accuracy analysis, the precision of segmentation is improved;
If the area of a certain ground block message is more than another plot, the segmentation precision corresponding to the plot of small area is improved;If a certainly
The relevance of block message and remaining ground block message is very high, then the segmentation precision of the ground block message can be improved, or be based on its leeway
Segmentation precision corresponding to block message, to determine the segmentation precision etc. in current plot.
It preferably, can be in conjunction with the geographical location corresponding to described ground block message when obtaining the Raster Images data
Information determines the information such as the segmentation precision of extraction Raster Images data, thus based on identified processing mode, it is described to obtain
Raster Images data corresponding to ground block message.
In step s3, the interpretation device extracts the characteristics of image in the Raster Images data.
Specifically, in step s3, the interpretation device carries out feature for each acquired Raster Images data
Extraction, to obtain the characteristics of image in the Raster Images data.Those skilled in the art will be understood that when the different figure of extraction
When as feature, corresponding extracting method is different.
Preferably, in step s3, the interpretation device can be according to the original wave corresponding to the Raster Images data
Segment information extracts the characteristics of image in the Raster Images data.
Specifically, in step s3, the interpretation device can to the original band class informations of the Raster Images data into
Row processing, and described image feature is extracted from original band class information.
It is highly preferred that described image feature includes following at least any one:
The derivation information of the original band class information of the Raster Images data and the original band class information:The derivative
Information include but not limited to the original band class information is normalized, binaryzation or color space conversion etc. it is one or more
After processing, the information that is obtained.
The index information of the Raster Images data:Calculated using the original band combination of the Raster Images data
The index information gone out.The index information includes but not limited to:NDMI, i.e. difference humidity index (vegetation water content);NDVI, i.e.,
Normalized differential vegetation index;ARVI, i.e. air impedance vegetation index;NBR, i.e. normalization burning ratio;NBR2, i.e. normalization burning
Ratio 2;EVI, i.e. environmental vegetation index;SR, i.e. ratio vegetation index etc..
The texture information of the Raster Images data:The texture information mainly include Raster Images data adjacent picture elements it
Between relationship, i.e., feature the pixel extracted by modes such as convolution, ponds between Raster Images data.
In step s 4, the interpretation device determines the earth's surface corresponding to described ground block message according to described image feature
Type.
Specifically, in step s 4, the interpretation device utilizes intelligent algorithm, and described image characteristic use is advance
Trained model is classified, by the way that described image feature to be compared with ground surface type, with determination described ground block message institute
Corresponding ground surface type.
Fig. 4 shows a kind of method stream of automatic interpretation for remote sensing image according to a preferred embodiment of the present invention
Cheng Tu.
Specifically, in step sl, the interpretation device obtains one or more of remote sensing image ground block message in step
In rapid S2, the interpretation device obtains the Raster Images data corresponding to described ground block message;In step s3, the interpretation is set
The standby characteristics of image extracted in the Raster Images data;In step S41, the interpretation device according to described image feature,
Determine the candidate ground surface type of one or more corresponding to described ground block message;In step S42, the interpretation device is according to institute
Candidate ground surface type is stated, the ground surface type information in the one or more associations plot in conjunction with corresponding to described ground block message determines
Ground surface type corresponding to described ground block message.
Wherein, the step S1, the step S2 and the step S3 is identical as the correspondence step described in Fig. 3 or phase
Seemingly, therefore details are not described herein, and is incorporated herein by reference.
In step S41, the interpretation device determines one corresponding to described ground block message according to described image feature
Or multiple candidate ground surface types.
Specifically, in step S41, the interpretation device utilizes intelligent algorithm, and described image characteristic use is pre-
First trained model is classified, by the way that described image feature to be compared with ground surface type, with determination described ground block message
The candidate ground surface type ground surface type of corresponding one or more.
Preferably, in step S41, the interpretation device can be according to described image feature, in conjunction with scheduled classification mould
Type determines the candidate ground surface type of one or more corresponding to described ground block message, wherein candidate's ground surface type includes waiting
Selection of land table typonym and corresponding candidate ground surface type probability.
Specifically, the disaggregated model includes but not limited to vector machine, convolutional neural networks, random forest, linear classification
Device etc., in obtained candidate's ground surface type, including the title of candidate ground surface type and corresponding candidate ground surface type probability.
Here, according to a kind of disaggregated model, then the candidate ground surface type title and the candidate ground surface type probability are directly corresponding;
According to a variety of disaggregated models, then the probability corresponding to same candidate ground surface type title that a variety of disaggregated models can be calculated
Be weighted or weighted average after, determine the candidate ground surface type probability corresponding to candidate's ground surface type title.
In step S42, the interpretation device is according to the candidate ground surface type, in conjunction with corresponding to described ground block message
The ground surface type information in one or more associations plot, determines the ground surface type corresponding to described ground block message.
In step S42, one or more associations plot institute of the interpretation device corresponding to the block message of current position
Corresponding ground surface type information is selected in conjunction with the candidate ground surface type in the current plot from the candidate ground surface type
Go out the ground surface type corresponding to described ground block message.Wherein, the association plot includes but not limited to being adjacent to for current plot
The ground surface type information in the similar plot of block, shape, the similar plot of feature etc., the association plot has been identified or
Through obtaining multiple candidate ground surface type information, to which the interpretation device is according to the ground surface type information for being associated with plot
Carry out from the candidate ground surface type in current plot selection and confirm ground surface type, or by the candidate ground surface type in the association plot
In, highest one of probability as reference, with the ground surface type in the current plot of determination.
Here, the ground surface type in the association plot can be identical as the ground surface type of current position block message or not
Together, for example, the ground surface type in plot will can be directly associated with as the ground surface type of the current position block message, alternatively, for example,
Can be facing-port industry area by adjacent building Parcel division if the ground surface type for being associated with plot is ocean.
Preferably, in step S42, the interpretation device can be according between described ground block message and other ground block messages
Position incidence relation, determine the one or more associations plot corresponding to described ground block message;According to the candidate earth's surface class
Type determines the ground surface type corresponding to described ground block message in conjunction with the ground surface type information in the association plot.
Specifically, in step S42, the interpretation device can be according between described ground block message and other ground block messages
Position incidence relation, will be with plot of the position range of current position block message in certain threshold value as being associated with plot.Here,
The position range include but not limited at a distance from the current position block message in certain threshold value, or be such as in same
On the relative positions such as latitude or same longitude;Further, the reference factors such as administrative division are can be combined with, to determine association
Plot.
Then, in step S42, the interpretation device is by the rule of formulation, according to the candidate ground surface type, in conjunction with
The ground surface type information in the association plot, determines the ground surface type corresponding to described ground block message.If being associated with the earth's surface in plot
Type is ocean, then can be facing-port industry area by adjacent building Parcel division.
It should be noted that the present invention can be carried out in the assembly of software and/or software and hardware, for example, can adopt
With application-specific integrated circuit (ASIC), general purpose computer or any other realized similar to hardware device.In one embodiment
In, software program of the invention can be executed by processor to realize steps described above or function.Similarly, of the invention
Software program (including relevant data structure) can be stored in computer readable recording medium storing program for performing, for example, RAM memory,
Magnetic or optical driver or floppy disc and similar devices.In addition, hardware can be used to realize in some steps or function of the present invention, example
Such as, coordinate to execute the circuit of each step or function as with processor.
In addition, the part of the present invention can be applied to computer program product, such as computer program instructions, when its quilt
When computer executes, by the operation of the computer, it can call or provide according to the method for the present invention and/or technical solution.
And the program instruction of the method for the present invention is called, it is possibly stored in fixed or moveable recording medium, and/or pass through
Broadcast or the data flow in other signal loaded mediums and be transmitted, and/or be stored according to described program instruction operation
In the working storage of computer equipment.Here, including a device according to one embodiment of present invention, which includes using
Memory in storage computer program instructions and processor for executing program instructions, wherein when the computer program refers to
When order is executed by the processor, method and/or skill of the device operation based on aforementioned multiple embodiments according to the present invention are triggered
Art scheme.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case of without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Profit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent requirements of the claims
Variation includes within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.This
Outside, it is clear that one word of " comprising " is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in device claim is multiple
Unit or device can also be realized by a unit or device by software or hardware.The first, the second equal words are used for table
Show title, and does not represent any particular order.
Claims (16)
1. a kind of method of automatic interpretation for remote sensing image, wherein this approach includes the following steps:
Obtain one or more of remote sensing image ground block message;
Obtain the Raster Images data corresponding to described ground block message;
Extract the characteristics of image in the Raster Images data;
According to described image feature, the ground surface type corresponding to described ground block message is determined.
2. according to the method described in claim 1, wherein, the step of obtaining one or more of remote sensing image ground block message, wraps
It includes:
Image preprocessing is carried out to remote sensing images, to generate remote sensing images to be detected;
Straight-line detection is carried out to the remote sensing image to be detected, to extract one or more on the remote sensing images to be detected
Straight line;
Linear optimal processing is carried out to the straight line, optimizes straight line to determine;
According to the optimization straight line, the profile in the remote sensing images is extracted;
The profile extracted is fitted, to generate one or more ground block message in the remote sensing images.
3. method according to claim 1 or 2, wherein obtain the Raster Images data corresponding to described ground block message
Step includes:
According to described ground block message, operation is split to the remote sensing image corresponding to described ground block message, it is described to obtain
Raster Images data corresponding to block message.
4. according to the method in any one of claims 1 to 3, wherein the image extracted in the Raster Images data is special
The step of sign includes:
According to the original band class information corresponding to the Raster Images data, the image extracted in the Raster Images data is special
Sign.
5. method according to claim 1 to 4, wherein described image feature includes following at least any one:
The derivation information of the original band class information of the Raster Images data and the original band class information;
The index information of the Raster Images data;
The texture information of the Raster Images data.
6. the method according to any one of claims 1 to 5, wherein it is determined that earth's surface class corresponding to described ground block message
The step of type includes:
According to described image feature, the candidate ground surface type of one or more corresponding to described ground block message is determined;
According to the candidate ground surface type, the ground surface type in the one or more associations plot in conjunction with corresponding to described ground block message
Information determines the ground surface type corresponding to described ground block message.
7. according to the method described in claim 6, wherein it is determined that candidate earth's surfaces of one or more corresponding to described ground block message
The step of type includes:
According to described image feature, in conjunction with scheduled disaggregated model, one or more times corresponding to described ground block message are determined
Select ground surface type, wherein candidate's ground surface type includes that candidate ground surface type title and corresponding candidate ground surface type are general
Rate.
8. the method described according to claim 6 or 7, wherein in conjunction with the position related information between multiple plot to determine
State ground block message corresponding to ground surface type the step of include:
According to the position incidence relation between described ground block message and other ground block messages, determine corresponding to described ground block message
One or more associations plot;
According to the candidate ground surface type, in conjunction with the ground surface type information in the association plot, determine that described ground block message institute is right
The ground surface type answered.
9. a kind of interpretation device of automatic interpretation for remote sensing image, wherein the interpretation device includes:
First device, for obtaining one or more of remote sensing image ground block message;
Second device, for obtaining the Raster Images data corresponding to described ground block message;
3rd device, for extracting the characteristics of image in the Raster Images data;
4th device, for according to described image feature, determining the ground surface type corresponding to described ground block message.
10. interpretation device according to claim 9, wherein the first device is used for:
Image preprocessing is carried out to remote sensing images, to generate remote sensing images to be detected;
Straight-line detection is carried out to the remote sensing image to be detected, to extract one or more on the remote sensing images to be detected
Straight line;
Linear optimal processing is carried out to the straight line, optimizes straight line to determine;
According to the optimization straight line, the profile in the remote sensing images is extracted;
The profile extracted is fitted, to generate one or more ground block message in the remote sensing images.
11. interpretation device according to claim 9 or 10, wherein the second device is used for:
According to described ground block message, operation is split to the remote sensing image corresponding to described ground block message, it is described to obtain
Raster Images data corresponding to block message.
12. the interpretation device according to any one of claim 9 to 11, wherein the 3rd device is used for:
According to the original band class information corresponding to the Raster Images data, the image extracted in the Raster Images data is special
Sign.
13. the interpretation device according to any one of claim 9 to 12, wherein described image feature include it is following at least
Any one:
The derivation information of the original band class information of the Raster Images data and the original band class information;
The index information of the Raster Images data;
The texture information of the Raster Images data.
14. the interpretation device according to any one of claim 9 to 13, wherein the 4th device includes:
Unit 41, for according to described image feature, determining the candidate earth's surface of one or more corresponding to described ground block message
Type;
Unit four or two are used for according to the candidate ground surface type, the one or more associations in conjunction with corresponding to described ground block message
The ground surface type information in plot, determines the ground surface type corresponding to described ground block message.
15. interpretation device according to claim 14, wherein Unit 41 is used for:
According to described image feature, in conjunction with scheduled disaggregated model, one or more times corresponding to described ground block message are determined
Select ground surface type, wherein candidate's ground surface type includes that candidate ground surface type title and corresponding candidate ground surface type are general
Rate.
16. the interpretation device according to claims 14 or 15, wherein Unit four or two is used for:
According to the position incidence relation between described ground block message and other ground block messages, determine corresponding to described ground block message
One or more associations plot;
According to the candidate ground surface type, in conjunction with the ground surface type information in the association plot, determine that described ground block message institute is right
The ground surface type answered.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810007957.7A CN108376232A (en) | 2018-01-04 | 2018-01-04 | A kind of method and apparatus of automatic interpretation for remote sensing image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810007957.7A CN108376232A (en) | 2018-01-04 | 2018-01-04 | A kind of method and apparatus of automatic interpretation for remote sensing image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108376232A true CN108376232A (en) | 2018-08-07 |
Family
ID=63015559
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810007957.7A Withdrawn CN108376232A (en) | 2018-01-04 | 2018-01-04 | A kind of method and apparatus of automatic interpretation for remote sensing image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108376232A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109190497A (en) * | 2018-08-09 | 2019-01-11 | 成都天地量子科技有限公司 | A kind of arable land recognition methods based on the multispectral satellite image of timing |
CN109767488A (en) * | 2019-01-23 | 2019-05-17 | 广东康云科技有限公司 | Three-dimensional modeling method and system based on artificial intelligence |
CN110059758A (en) * | 2019-04-24 | 2019-07-26 | 海南长光卫星信息技术有限公司 | A kind of remote sensing image culture pond detection method based on semantic segmentation |
CN110070182A (en) * | 2019-04-01 | 2019-07-30 | 京微齐力(北京)科技有限公司 | The platform chip of suitable artificial intelligence and its manufacture and design method |
CN110298366A (en) * | 2019-07-05 | 2019-10-01 | 北华航天工业学院 | Crops are distributed extracting method and device |
CN110322546A (en) * | 2019-05-14 | 2019-10-11 | 广东康云科技有限公司 | Substation's three-dimensional digital modeling method, system, device and storage medium |
CN110322545A (en) * | 2019-05-14 | 2019-10-11 | 广东康云科技有限公司 | Campus three-dimensional digital modeling method, system, device and storage medium |
CN110570427A (en) * | 2019-07-19 | 2019-12-13 | 武汉珈和科技有限公司 | Remote sensing image semantic segmentation method and device fusing edge detection |
CN111126298A (en) * | 2019-12-25 | 2020-05-08 | 二十一世纪空间技术应用股份有限公司 | Interpretation method and device of remote sensing information and electronic equipment |
CN111382645A (en) * | 2018-12-29 | 2020-07-07 | 顺丰科技有限公司 | Method and system for identifying expired buildings in electronic map |
CN113642444A (en) * | 2021-08-06 | 2021-11-12 | 苏州九宇遥感科技有限公司 | Automatic interpretation method for land types based on low-frequency error elimination |
CN113763396A (en) * | 2021-09-02 | 2021-12-07 | 中国农业科学院农业信息研究所 | Remote sensing image plot extraction method and system based on deep learning |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101840582A (en) * | 2010-02-05 | 2010-09-22 | 北京交通大学 | Boundary digitizing method of cadastral plot |
CN103364781A (en) * | 2012-04-11 | 2013-10-23 | 南京财经大学 | Remote sensing data and geographical information system-based grainfield ground reference point screening method |
CN105303184A (en) * | 2015-11-25 | 2016-02-03 | 中国矿业大学(北京) | Method for accurately identifying ground features in satellite remote-sensing image |
CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
CN106503739A (en) * | 2016-10-31 | 2017-03-15 | 中国地质大学(武汉) | The target in hyperspectral remotely sensed image svm classifier method and system of combined spectral and textural characteristics |
CN107092930A (en) * | 2017-04-21 | 2017-08-25 | 中国科学院遥感与数字地球研究所 | It is a kind of by DIGITAL PLANNING map(DLG)Data are used for the method that high-resolution remote sensing image ground mulching is classified |
CN107392926A (en) * | 2017-09-18 | 2017-11-24 | 河海大学 | Characteristics of remote sensing image system of selection based on soil thematic map early stage |
-
2018
- 2018-01-04 CN CN201810007957.7A patent/CN108376232A/en not_active Withdrawn
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101840582A (en) * | 2010-02-05 | 2010-09-22 | 北京交通大学 | Boundary digitizing method of cadastral plot |
CN103364781A (en) * | 2012-04-11 | 2013-10-23 | 南京财经大学 | Remote sensing data and geographical information system-based grainfield ground reference point screening method |
CN105303184A (en) * | 2015-11-25 | 2016-02-03 | 中国矿业大学(北京) | Method for accurately identifying ground features in satellite remote-sensing image |
CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
CN106503739A (en) * | 2016-10-31 | 2017-03-15 | 中国地质大学(武汉) | The target in hyperspectral remotely sensed image svm classifier method and system of combined spectral and textural characteristics |
CN107092930A (en) * | 2017-04-21 | 2017-08-25 | 中国科学院遥感与数字地球研究所 | It is a kind of by DIGITAL PLANNING map(DLG)Data are used for the method that high-resolution remote sensing image ground mulching is classified |
CN107392926A (en) * | 2017-09-18 | 2017-11-24 | 河海大学 | Characteristics of remote sensing image system of selection based on soil thematic map early stage |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109190497A (en) * | 2018-08-09 | 2019-01-11 | 成都天地量子科技有限公司 | A kind of arable land recognition methods based on the multispectral satellite image of timing |
CN111382645B (en) * | 2018-12-29 | 2023-11-28 | 顺丰科技有限公司 | Method and system for identifying overdue building in electronic map |
CN111382645A (en) * | 2018-12-29 | 2020-07-07 | 顺丰科技有限公司 | Method and system for identifying expired buildings in electronic map |
CN109767488A (en) * | 2019-01-23 | 2019-05-17 | 广东康云科技有限公司 | Three-dimensional modeling method and system based on artificial intelligence |
CN110070182A (en) * | 2019-04-01 | 2019-07-30 | 京微齐力(北京)科技有限公司 | The platform chip of suitable artificial intelligence and its manufacture and design method |
CN110070182B (en) * | 2019-04-01 | 2021-08-24 | 京微齐力(北京)科技有限公司 | Platform chip suitable for artificial intelligence and manufacturing and designing method thereof |
CN110059758A (en) * | 2019-04-24 | 2019-07-26 | 海南长光卫星信息技术有限公司 | A kind of remote sensing image culture pond detection method based on semantic segmentation |
CN110322546A (en) * | 2019-05-14 | 2019-10-11 | 广东康云科技有限公司 | Substation's three-dimensional digital modeling method, system, device and storage medium |
CN110322545A (en) * | 2019-05-14 | 2019-10-11 | 广东康云科技有限公司 | Campus three-dimensional digital modeling method, system, device and storage medium |
CN110298366A (en) * | 2019-07-05 | 2019-10-01 | 北华航天工业学院 | Crops are distributed extracting method and device |
CN110298366B (en) * | 2019-07-05 | 2021-05-04 | 北华航天工业学院 | Crop distribution extraction method and device |
CN110570427A (en) * | 2019-07-19 | 2019-12-13 | 武汉珈和科技有限公司 | Remote sensing image semantic segmentation method and device fusing edge detection |
CN111126298A (en) * | 2019-12-25 | 2020-05-08 | 二十一世纪空间技术应用股份有限公司 | Interpretation method and device of remote sensing information and electronic equipment |
CN111126298B (en) * | 2019-12-25 | 2024-01-09 | 二十一世纪空间技术应用股份有限公司 | Remote sensing information interpretation method and device and electronic equipment |
CN113642444A (en) * | 2021-08-06 | 2021-11-12 | 苏州九宇遥感科技有限公司 | Automatic interpretation method for land types based on low-frequency error elimination |
CN113763396A (en) * | 2021-09-02 | 2021-12-07 | 中国农业科学院农业信息研究所 | Remote sensing image plot extraction method and system based on deep learning |
CN113763396B (en) * | 2021-09-02 | 2023-07-25 | 中国农业科学院农业信息研究所 | Remote sensing image land parcel extraction method and system based on deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108376232A (en) | A kind of method and apparatus of automatic interpretation for remote sensing image | |
US9292747B2 (en) | Methods and systems for automatic and semi-automatic geometric and geographic feature extraction | |
CN109191432B (en) | Remote sensing image cloud detection method based on domain transformation filtering multi-scale decomposition | |
CN110717489A (en) | Method and device for identifying character area of OSD (on screen display) and storage medium | |
CN112465889B (en) | Plant point cloud segmentation method, system and storage medium based on two-dimensional-three-dimensional integration | |
CN107025687B (en) | A kind of view-based access control model relies on the historical relic line chart method for drafting of Curvature Estimate | |
US11804025B2 (en) | Methods and systems for identifying topographic features | |
CN103971377A (en) | Building extraction method based on prior shape level set segmentation | |
CN107992856A (en) | High score remote sensing building effects detection method under City scenarios | |
CN110751606A (en) | Foam image processing method and system based on neural network algorithm | |
CN111260564A (en) | Image processing method and device and computer storage medium | |
CN106599891A (en) | Remote sensing image region-of-interest rapid extraction method based on scale phase spectrum saliency | |
Zheng et al. | Single shot multibox detector for urban plantation single tree detection and location with high-resolution remote sensing imagery | |
Sapirstein | Segmentation, reconstruction, and visualization of ancient inscriptions in 2.5 D | |
Kim et al. | Generation of a DTM and building detection based on an MPF through integrating airborne lidar data and aerial images | |
Zhang et al. | Automatic extraction of building geometries based on centroid clustering and contour analysis on oblique images taken by unmanned aerial vehicles | |
Omidalizarandi et al. | Segmentation and classification of point clouds from dense aerial image matching | |
CN109117841B (en) | Scene text detection method based on stroke width transformation and convolutional neural network | |
Liu et al. | Automated building change detection using UltraCamD images and existing CAD data | |
Cheng et al. | Power pole detection based on graph cut | |
CN110598702A (en) | Preparation method, system, equipment and medium of ship target detection data set | |
Maltezos et al. | Improving the visualisation of 3D textured models via shadow detection and removal | |
Fernandes et al. | Extraction of building roof contours from the integration of high-resolution aerial imagery and laser data using Markov random fields | |
CN115880325A (en) | Building outline automatic extraction method based on point cloud dimension and spatial distance clustering | |
Guili et al. | A man-made object detection algorithm based on contour complexity evaluation |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180807 |