CN105224938A - The modification method of remote sensing images land use classes result - Google Patents
The modification method of remote sensing images land use classes result Download PDFInfo
- Publication number
- CN105224938A CN105224938A CN201510753837.8A CN201510753837A CN105224938A CN 105224938 A CN105224938 A CN 105224938A CN 201510753837 A CN201510753837 A CN 201510753837A CN 105224938 A CN105224938 A CN 105224938A
- Authority
- CN
- China
- Prior art keywords
- land
- classification
- result
- certain time
- remote sensing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to image processing field, provide a kind of modification method of remote sensing images land use classes result, the method comprises definition solution space, utilizes the method being suitable for searching for organize solution space, utilize depth-first traversal solution space.The technical scheme that the present invention proposes more is conducive to the correction of the low classification of nicety of grading, and adjustment in accuracy is according to being conducive to the lower direction running of automatic recognition classification precision, is therefore more conducive to the classify accuracy balancing each atural object classification.
Description
Technical field
The present invention relates to image processing field, particularly a kind of modification method of remote sensing images land use classes result.
Background technology
Land_use change is often referred to the different attribute of the mankind for different soil, in order to economy and the object of society, adopt various technological means, in soil enterprising line period or long-term various activities reach needed for the mankind.It is the overriding concern factor of carrying out scientific management and decision-making in resources and environmental protection that Land Use/land covers (LandUseandLandCover, be called for short LULC) Changeement.Utilizing remotely-sensed data to carry out the variation monitoring research of Land_use change and land cover pattern, there is following shortcoming in the object oriented classification result data based on fuzzy mathematics thought:
(1) the impact of " the different spectrum of jljl, foreign matter is with spectrum " phenomenon, atural object classification border mixed pixel is more, makes the partitioning boundary of some types of ground objects undesirable, often classifies inaccurate by the result that algorithm extracts;
(2) there is " salt-pepper noise " phenomenon, existing of broken thin objects is outdated, causes and produce the less object of fragmentary distribution area on the same atural object in same region;
(3) the impact of shade.Often there is shade in the back on mountain range, in brightness value with average all with classify not too consistent under normal circumstances, affect the effect of automatic classification;
(4), there are some mistakes unavoidably and divide or leak the phenomenon of dividing in large area automatic classification.
At present, in order to improve classification progress, sorting technique can only be changed.
Summary of the invention
[technical matters that will solve]
The object of this invention is to provide a kind of modification method of remote sensing images land use classes result, go out to send correction nicety of grading from this angle of classification results.
[technical scheme]
The present invention is achieved by the following technical solutions.
The present invention relates to a kind of modification method of remote sensing images land use classes result, the method comprising the steps of:
A, definition solution space, described solution space comprises the type of ground objects in the terrain classification result of remote sensing images;
B, organize solution space according to the land classification result of Land_use change and land cover dynamic Changing Pattern and different year, the land classification result of described different year comprises land classification result as a certain time of modified basis and the land classification result in a certain time to be revised, and described solution space comprises the possible variation tendency of each type of ground objects in each time;
C, searching route is set according to the solution space organized in step B and according to searching route ergodic solutions space, the method in described ergodic solutions space is: from the land classification result in a certain time as modified basis, select a type of ground objects, the classification results treating a certain time of correction one by one carries out searching for and marks the object searched, and repeats this operation until all selected as all types of ground objects in the land classification result in a certain time of modified basis;
D, OO remote sensing image fuzzy classifier method is adopted to classify to the object marked in step C, revise the object of classification error in the classification results in a certain time to be revised, and give the classification results in a certain time to be revised revised result and obtain final classification results.
As one preferred embodiment, described type of ground objects comprises forest land, meadow, wetland, arable land, artificial surface, unused land.
[beneficial effect]
The technical scheme that the present invention proposes has following beneficial effect:
Backtracking modification method of the present invention is more conducive to the correction of the low classification of nicety of grading, and adjustment in accuracy is according to being conducive to the lower direction running of automatic recognition classification precision, therefore more can also be conducive to the classify accuracy balancing each atural object classification.
Accompanying drawing explanation
Fig. 1 is the backtracking disaggregation figure of the embodiment of the present invention.
Fig. 2 is the backtracking traverse path figure of the embodiment of the present invention.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly, carry out clear, complete description by the specific embodiment of the present invention below.
Embodiment
The embodiment of the present invention is revised remote sensing images land use classes result based on the present invention is based on backtracking method, backtracking method is also commonly called as trial method, problem candidate solution is checked by setting order and is exemplified by successively, if candidate solution does not satisfy condition, then abandon this candidate solution, and select next retrieval direction; If this candidate solution relieves also do not meet current needs, just expand current candidate solution scale, continued checking.If this candidate solution meets all conditions comprising problem scale, this candidate solution is exactly required solution.In this process, the candidate solution do not satisfied condition is removed, selects the flow process of next candidate solution to be just called backtracking, expand current candidate solution scale, the exploration process having continued checking is called that forward direction is soundd out.
Particularly, in the present embodiment, suppose the land use classes figure having obtained now nineteen ninety, 2000 and 2010, need now to utilize the land use classes figure of the land use classes figure of 2010 to nineteen ninety, 2000 to revise.Modification method comprises definition solution space, utilizes the method being suitable for searching for organize solution space, utilize depth-first traversal solution space.Respectively each step of modification method is described in detail below.
First solution space is defined.Particularly, according to the terrain classification result of remote sensing images, the present embodiment have employed Chinese Academy of Sciences's land use classification system, is divided into six parts: forest land, meadow, wetland, and plough, artificial surface, unused land, so solution space is defined as this 6 class atural object.
Then the method being suitable for searching for is utilized to organize solution space.Particularly, according to the Changing Pattern that Land_use change and land cover dynamic change, the land use classes figure of different year contrasts, by analyzing the change of often kind of atural object, to be exactly the atural object of analyzing change be specifically by which kind of feature changes original formed organize solution space, as shown in Figure 1.
Secondly, searching route is set.As shown in Figure 2, ImagesPos (image traversal pointer) searches for nineteen ninety classification results data one by one according to the class categories of 2010, direction is-90 years forest lands, 10 years forest lands,-90 years meadows, 10 years forest lands,-90 years wetlands in 10 years forest lands, plough in 10 years forest lands for-90 years ,-90 years artificial surfaces in 10 years forest lands, 10 years forest lands-90 years other.Then, change 10 years atural object classifications, then 90 years atural object is traveled through, the like, reach whole solution space traversal.Particularly, travel through whole area image according to Fig. 2, and the object that demarcation searches is 10**-90**, is 10 forest lands as searched object, 90 meadows, then demarcating object is 10 meadows, forest land-90, by that analogy.Due to traversal tagged object be all can very clear resolution be whether the object of corresponding atural object, so again adopt OO remote sensing image fuzzy classifier method to classify to tagged object, just can revise the object by mistake divided with wrong point, again revised result is given on 90 years classification images, just obtain final correction result.
The precision change that front and back are revised in backtracking is as shown in table 1.
Table 1 precision evaluation statistical form
After revising front and correction from backtracking, precision result compares, wetland adjustment in accuracy amplitude is slightly low, unused land precision increase rate is maximum, this illustrates that the backtracking modification method in the embodiment of the present invention is more conducive to the correction of the low classification of nicety of grading, and adjustment in accuracy is according to being conducive to the lower direction running of automatic recognition classification precision, is therefore more conducive to the classify accuracy balancing each atural object classification.
Need to illustrate, the embodiment of foregoing description is a part of embodiment of the present invention, instead of whole embodiment, neither limitation of the present invention.Based on embodiments of the invention, those of ordinary skill in the art, not paying the every other embodiment obtained under creative work prerequisite, belong to protection scope of the present invention.
Claims (2)
1. a modification method for remote sensing images land use classes result, is characterized in that comprising step:
A, definition solution space, described solution space comprises the type of ground objects in the terrain classification result of remote sensing images;
B, organize solution space according to the land classification result of Land_use change and land cover dynamic Changing Pattern and different year, the land classification result of described different year comprises land classification result as a certain time of modified basis and the land classification result in a certain time to be revised, and described solution space comprises the possible variation tendency of each type of ground objects in each time;
C, searching route is set according to the solution space organized in step B and according to searching route ergodic solutions space, the method in described ergodic solutions space is: from the land classification result in a certain time as modified basis, select a type of ground objects, the classification results treating a certain time of correction one by one carries out searching for and marks the object searched, and repeats this operation until all selected as all types of ground objects in the land classification result in a certain time of modified basis;
D, OO remote sensing image fuzzy classifier method is adopted to classify to the object marked in step C, revise the object of classification error in the classification results in a certain time to be revised, and give the classification results in a certain time to be revised revised result and obtain final classification results.
2. the modification method of remote sensing images land use classes result according to claim 1, is characterized in that described type of ground objects comprises forest land, meadow, wetland, arable land, artificial surface, unused land.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510753837.8A CN105224938B (en) | 2015-11-06 | 2015-11-06 | The modification method of remote sensing images land use classes result |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510753837.8A CN105224938B (en) | 2015-11-06 | 2015-11-06 | The modification method of remote sensing images land use classes result |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105224938A true CN105224938A (en) | 2016-01-06 |
CN105224938B CN105224938B (en) | 2018-11-09 |
Family
ID=54993898
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510753837.8A Active CN105224938B (en) | 2015-11-06 | 2015-11-06 | The modification method of remote sensing images land use classes result |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105224938B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112070078A (en) * | 2020-11-16 | 2020-12-11 | 武汉思众空间信息科技有限公司 | Deep learning-based land utilization classification method and system |
CN114581725A (en) * | 2022-05-06 | 2022-06-03 | 武汉光谷信息技术股份有限公司 | Ground feature classification method integrating multi-source data encryption and theoretical derivation |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100332475A1 (en) * | 2009-06-25 | 2010-12-30 | University Of Tennessee Research Foundation | Method and apparatus for predicting object properties and events using similarity-based information retrieval and modeling |
CN102289678A (en) * | 2011-08-10 | 2011-12-21 | 武汉大学 | Fuzzy supervised classification method for multiband remote sensing image based on non-equal weight distances |
CN104239890A (en) * | 2014-08-12 | 2014-12-24 | 浙江工商大学 | Method for automatically extracting coastal land and earth cover information through GF-1 satellite |
CN105005782A (en) * | 2014-04-24 | 2015-10-28 | 中国科学院遥感与数字地球研究所 | Fine method for global vegetation classification based on multi-temporal remote sensing data and spectroscopic data |
-
2015
- 2015-11-06 CN CN201510753837.8A patent/CN105224938B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100332475A1 (en) * | 2009-06-25 | 2010-12-30 | University Of Tennessee Research Foundation | Method and apparatus for predicting object properties and events using similarity-based information retrieval and modeling |
CN102289678A (en) * | 2011-08-10 | 2011-12-21 | 武汉大学 | Fuzzy supervised classification method for multiband remote sensing image based on non-equal weight distances |
CN105005782A (en) * | 2014-04-24 | 2015-10-28 | 中国科学院遥感与数字地球研究所 | Fine method for global vegetation classification based on multi-temporal remote sensing data and spectroscopic data |
CN104239890A (en) * | 2014-08-12 | 2014-12-24 | 浙江工商大学 | Method for automatically extracting coastal land and earth cover information through GF-1 satellite |
Non-Patent Citations (2)
Title |
---|
陈静谊: "黄土高原清水河流域植被景观格局变化与水土流失的影响研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
韦玮等: "基于偏差平均值的乌梁素海湿地变化检测研究", 《林业科学研究》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112070078A (en) * | 2020-11-16 | 2020-12-11 | 武汉思众空间信息科技有限公司 | Deep learning-based land utilization classification method and system |
CN112070078B (en) * | 2020-11-16 | 2021-01-19 | 武汉思众空间信息科技有限公司 | Deep learning-based land utilization classification method and system |
CN114581725A (en) * | 2022-05-06 | 2022-06-03 | 武汉光谷信息技术股份有限公司 | Ground feature classification method integrating multi-source data encryption and theoretical derivation |
Also Published As
Publication number | Publication date |
---|---|
CN105224938B (en) | 2018-11-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11836976B2 (en) | Method for recognizing seawater polluted area based on high-resolution remote sensing image and device | |
CN108846352B (en) | Vegetation classification and identification method | |
Gupta et al. | Object based information extraction from high resolution satellite imagery using eCognition | |
CN109409400A (en) | Merge density peaks clustering method, image segmentation system based on k nearest neighbor and multiclass | |
CN107330422A (en) | A kind of method for carrying out mima type microrelief classification to semiarid zone based on high accuracy number elevation model | |
Chen et al. | Optimal segmentation of a high-resolution remote-sensing image guided by area and boundary | |
CN109063730A (en) | A kind of convex programming cluster water pollution source tracing method | |
CN103888541A (en) | Method and system for discovering cells fused with topology potential and spectral clustering | |
CN109829425B (en) | Farmland landscape small-scale ground feature classification method and system | |
CN104657980A (en) | Improved multi-channel image partitioning algorithm based on Meanshift | |
CN108681715A (en) | Hainan Tropical wildwood vegetation type group sorting technique | |
CN107330027A (en) | A kind of Weakly supervised depth station caption detection method | |
CN106599915A (en) | Vehicle-mounted laser point cloud classification method | |
CN106228554A (en) | Fuzzy coarse central coal dust image partition methods based on many attribute reductions | |
CN104239890A (en) | Method for automatically extracting coastal land and earth cover information through GF-1 satellite | |
CN110458201A (en) | A kind of remote sensing image object-oriented classification method and sorter | |
CN108921025A (en) | A kind of object level classification samples automatic selecting method of collaborative variation detection | |
CN105868269A (en) | Precise image searching method based on region convolutional neural network | |
CN112989985B (en) | Urban built-up area extraction method integrating night light data and Landsat8OLI images | |
CN104182728A (en) | Vehicle logo automatic location and recognition method based on pattern recognition | |
CN106875481A (en) | A kind of preparation method of three-dimensional visualization remote sensing image Surface classification model | |
CN110147780A (en) | The landform recognition methods of real-time field robot and system based on level landform | |
CN108573238A (en) | A kind of vehicle checking method based on dual network structure | |
CN105224938A (en) | The modification method of remote sensing images land use classes result | |
CN107480634A (en) | A kind of geographical national conditions ground mulching monitoring method based on multistage target classification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |