CN105809189A - Time series image processing method - Google Patents
Time series image processing method Download PDFInfo
- Publication number
- CN105809189A CN105809189A CN201610119032.2A CN201610119032A CN105809189A CN 105809189 A CN105809189 A CN 105809189A CN 201610119032 A CN201610119032 A CN 201610119032A CN 105809189 A CN105809189 A CN 105809189A
- Authority
- CN
- China
- Prior art keywords
- time series
- classification
- image
- variation curve
- highlighted
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Abstract
The invention discloses a time series image processing method which is performed through the following steps: capturing a time series image; carrying out a smooth treatment to the time series image; capturing classified images according to the time series after the smooth treatment and a plurality of time series change curves corresponded by different classifications; displaying the classified images and the plurality of time series change curves corresponded by different classifications; According to the invention, by looking for the variation of space to time change in a time series image and by gathering plots with similar variation, classified images and time series change curves corresponded by different classifications can be classified and summed up to facilitate more users to browns the changed information of a time series image on the time series.
Description
Technical field
The invention belongs to network geographic information technical field, more particularly, to a kind of time series image treatment method based on WEBGIS technology.
Background technology
It is height correlation in the change in time series of grain security field, natural resources and weather with Grain Growth Situation.Using the remote sensing in time series or climatic data monitoring Grain Growth Situation change is general research method, but, in the digitized process of agricultural monitoring, how using network geographic information system (WEBGIS) technology is have very much challenge in page end monitoring and show image information in time series.
Traditional remote sense monitoring system many employings vector based on WEBGIS or the processing method of grid, this method is usually applicable to the image on single one time point of scape of display, shown the statistical data of diverse location information by vector data, use raster based method to show detailed Data distribution information.Grid cumulative desertification method for early warning (CN201210144411) in a kind of region utilizes the regional change of remote sensing image data monitoring desertification and its monitoring result is carried out comprehensive pre-warning.The WEBGIS system (CN201020558400) that a kind of arrow pattern and raster mode mixing use invents the WEBGIS system that vector combines with grid, in conjunction with WEBGIS technology for showing the remote sensing image of single time image.The technology that current remote sensing combines with WEBGIS is to be applied to all trades and professions, and WEBGIS technology can facilitate and uses the method that vector grid is combined to display the remote sensing image of single time point and statistical result intuitively, possesses intuitively, handy technical characterstic.But, seasonal effect in time series remote sensing images analysis is also traditional analysis and the using method of remotely-sensed data, cause adding in traditional simple X of image, Y coordinate the Z coordinate axle of time labelling owing to adding time series, be difficult to use traditional WEBGIS technology and display in webpage.
Summary of the invention
It is an object of the invention to provide a kind of time series image treatment method.
According to an aspect of the present invention, it is provided that a kind of time series image treatment method, including: obtain time series image；Described time series image is smoothed；The time series variation curve of image and the different classes of correspondence of classifying according to the described time series image capturing after smoothing processing；Show the time series variation curve of described classification image and different classes of correspondence.
Preferably, it is smoothed including to described time series image: divide pixel to process the image in time series；The exceptional value in time series image is eliminated according to pixel time series variation.
Preferably, time series variation curve according to the described time series image capturing classification image after smoothing processing and different classes of correspondence includes: by maximum likelihood classification and ISODATA clustering procedure, described time series image is divided, obtains the time series image of classification；Time series image capturing classification image according to described classification and the time series variation curve of different classes of correspondence.
Preferably, include according to the time series image capturing classification image of described classification and the time series variation curve of different classes of correspondence: according to the time changing curve of all pixels in each classification of time series image capturing of each classification；Each time point of all pixels in each classification is averaged and obtains the value of each time point in each classification；The time series variation curve of each classification is obtained according to the value of each time point in described each classification.
Preferably, described processing method also includes: obtains the pre-set categories in described classification image according to operational order, is highlighted by time series variation curve corresponding for described pre-set categories；Obtain the Preset Time sequence variation curve in the time series variation curve of different classes of correspondence according to operational order, classification corresponding for described Preset Time sequence variation curve is highlighted in described classification image.
Preferably, described classification image includes corresponding positional information or identification information with described time series variation curve.
Preferably, time series variation curve corresponding for described pre-set categories is highlighted includes: obtain the positional information of pre-set categories；The time series variation curve corresponding with described positional information is obtained according to described positional information；The time series variation curve corresponding with described positional information is highlighted.
Preferably, time series variation curve corresponding for described pre-set categories is highlighted includes: obtain the identification information of pre-set categories；The time series variation curve corresponding with described identification information is obtained according to described identification information；The time series variation curve corresponding with described identification information is highlighted.
Preferably, classification corresponding for described Preset Time sequence variation curve is highlighted in described classification image includes: obtain the positional information of Preset Time sequence variation curve；The classification corresponding with described positional information is obtained according to described positional information；The classification corresponding with described positional information is highlighted by described classification image.
Preferably, classification corresponding for described Preset Time sequence variation curve is highlighted in the image of described classification includes: obtain the identification information of Preset Time sequence variation curve；The classification corresponding with described identification information is obtained according to described identification information；The classification corresponding with described identification information is highlighted by described classification image.
Preferably, described processing method also includes: by described classification image store in the first data base；The time series variation curve of different classes of correspondence is stored in the second data base.
Preferably, the time series variation curve of described different classes of correspondence graphically shows.
Time series image treatment method provided by the invention, by finding different spaces difference on the time changes from time series image, assemble the plot of similar differences change, classify, collect the time series variation curve obtaining classification image and different classes of correspondence, facilitate more user's browsing time sequential images change information in time series.
Accompanying drawing explanation
By referring to the accompanying drawing description to the embodiment of the present invention, above-mentioned and other purposes of the present invention, feature and advantage will be apparent from, in the accompanying drawings:
Fig. 1 illustrates the flow chart of time series image treatment method according to embodiments of the present invention.
Fig. 2 illustrates the schematic diagram of time series image according to embodiments of the present invention；
Fig. 3 illustrates the schematic diagram of classification image according to embodiments of the present invention；
Fig. 4 illustrates the time series variation curve of different classes of correspondence according to embodiments of the present invention.
Detailed description of the invention
It is more fully described various embodiments of the present invention hereinafter with reference to accompanying drawing.In various figures, identical element adopts same or similar accompanying drawing labelling to represent.For the sake of clarity, the various piece in accompanying drawing is not necessarily to scale.
The present invention can present in a variety of manners, some of them example explained below.
Fig. 1 illustrates the flow chart of time series image treatment method according to embodiments of the present invention.As described in Figure 1, described time series image treatment method comprises the following steps.
In step S01, obtain time series image.
In the present embodiment, adopt the Moderate Imaging Spectroradiomete (moderate-resolutionimagingspectroradiometer, MODIS) in time series that the growing way during Crop in China phenology is changed to be monitored obtaining time series image.
MODIS can be used for earth's surface, biosphere, Solid Earth, air and ocean are carried out long-term global observation.
Growing way change during Crop in China phenology is monitored by NDVI (NormalizedDifferenceVegetationIndex, standard difference vegetation index) change.
In step S02, described time series image is smoothed.
In the present embodiment, step S02 comprises the following steps.
In step S021, pixel is divided to process the image in time series.
In step S022, eliminate the exceptional value in time series image according to pixel time series.
Specifically; when analysis time sequence remotely-sensed data; data usually can be subject to cloud snow or other effect of noise; change in pixel time series is produced impact; thus the analysis of influence time sequence curve, this example uses self adaptation Savitzky-Golay time series filtering method that the data in time series are smoothed.The method uses the filter on certain length that the data in time series are done convolution, thus time series data being done weighted polynomial matching, trying to achieve least squares error, the data away from most of points are not involved in matching, thus abandoning the noise section of deviation normal trend.Original series envelope is infinitely approached by the time-serial position after making matching after successive ignition.
On image after smoothing processing, the change in all crop pixel time serieses will not be subject to the interference of noise, and the time-serial position of all crop pixels is all close to its original series envelope.
In step S03, the time series variation curve of image and the different classes of correspondence of classifying according to the described time series image capturing after smoothing processing.
In the present embodiment, step S03 comprises the following steps.
In step S031, by maximum likelihood classification and ISODATA clustering procedure, described time series image is divided, obtain the time series image of classification.
In the present embodiment, by setting initial parameter and using the mechanism of merger and division, when certain two class cluster centre distance is less than predetermined threshold value, they are merged into same category, when certain quasi-standard deviation exceedes predetermined threshold value more than predetermined threshold value or its number of samples, it is classified as two classifications.When certain class number of samples is less than predetermined threshold value, need to be cancelled.So, according to parameter iterations such as the class numbers of initial cluster center and setting, a more satisfactory classification results is finally given.
In step S032, the time series variation curve of image and the different classes of correspondence of classifying according to the time series image capturing of described classification.
In the present embodiment, step S032 includes: according to the time changing curve of all pixels in each classification of time series image capturing of each classification；Each time point of all pixels in each classification is averaged and obtains the value of each time point in each classification；The time series variation curve of each classification is obtained according to the value of each time point in described each classification.
Specifically, by clustering method, time series image is divided the time changing curve obtaining all pixels of each classification, the curve of all pixels of each time point of each classification is averaged, obtain the value of each time point of each classification, thus obtaining the time series variation curve of each classification.
In step S04, show the time series variation curve of described classification image and different classes of correspondence.
In the present embodiment, by described classification image store in map data base, and different classes of classification image is issued into Map Services by cutting the mode of figure, be illustrated in webpage by WEGIS technology.The time series variation curve of different classes of correspondence is stored in diagram database simultaneously, calls for interactive graphical user interface E-CHART.The time series variation curve of described different classes of correspondence graphically shows.
In a preferred embodiment, described classification image includes corresponding positional information or identification information with described time series variation curve.Wherein, described positional information is the coordinate of the mark position of the category, the coordinate of the mark position that the time series variation curve of its correspondence is also respectively corresponding.Described identification information can be numbering, as being 1,2,3,4 by the class number of division ..., the numbering 1,2,3,4 that the time series variation curve of its correspondence is also corresponding respectively ....
Described time series image treatment method also includes step S05-step S07.
In step S05, obtain the pre-set categories in described classification image according to operational order, time series variation curve corresponding for described pre-set categories is highlighted.
In the present embodiment, step S05 includes: obtain the positional information of pre-set categories；The time series variation curve corresponding with described positional information is obtained according to described positional information；The time series variation curve corresponding with described positional information is highlighted.
Another preferred embodiment in, step S05 includes: obtain pre-set categories identification information；The time series variation curve corresponding with described identification information is obtained according to described identification information；The time series variation curve corresponding with described identification information is highlighted.
In step S06, obtain the Preset Time sequence variation curve in the time series variation curve of different classes of correspondence according to operational order, classification corresponding for described Preset Time sequence variation curve is highlighted in described classification image.
In the present embodiment, step S06 includes: obtain the positional information of Preset Time sequence variation curve；The classification corresponding with described positional information is obtained according to described positional information；The classification corresponding with described positional information is highlighted by described classification image.
Another preferred embodiment in, step S06 includes: obtain Preset Time sequence variation curve identification information；The classification corresponding with described identification information is obtained according to described identification information；The classification corresponding with described identification information is highlighted by described classification image.
Specifically, by JS Both Internet language, the broken line graph in E-CHART is got up with the Map Services linkage in WEBGIS, realize clicking classification chart picture in map, in its broken line graph, corresponding sorting track is highlighted, can real time inspection this classification of different periods time change, meanwhile, click event sequence curve, place cartographic classification also can be simultaneously highlighted, for determining the geographical position classified in place.
Time series image treatment method provided by the invention, by finding different spaces difference on the time changes from time series image, assemble the plot of similar differences change, classify, collect the time series variation curve obtaining classification image and different classes of correspondence, facilitate more user's browsing time sequential images change information in time series.
According to embodiments of the invention as described above, these embodiments do not have all of details of detailed descriptionthe, are not intended to the specific embodiment that this invention is only described yet.Obviously, as described above, can make many modifications and variations.These embodiments are chosen and specifically described to this specification, is to explain principles of the invention and practical application better, so that skilled artisan can utilize the present invention and the amendment on basis of the present invention to use well.Protection scope of the present invention should be as the criterion with the scope that the claims in the present invention define.
Claims (12)
1. a time series image treatment method, including:
Obtain time series image；
Described time series image is smoothed；
The time series variation curve of image and the different classes of correspondence of classifying according to the described time series image capturing after smoothing processing；
Show the time series variation curve of described classification image and different classes of correspondence.
2. processing method according to claim 1, wherein, is smoothed including to described time series image:
Pixel is divided to process the image in time series；
The exceptional value in time series image is eliminated according to pixel time series variation.
3. processing method according to claim 1, wherein, includes according to the time series variation curve of the described time series image capturing classification image after smoothing processing and different classes of correspondence:
By maximum likelihood classification and ISODATA clustering procedure, described time series image is divided, obtain the time series image of classification；
Time series image capturing classification image according to described classification and the time series variation curve of different classes of correspondence.
4. processing method according to claim 3, wherein, includes according to the time series variation curve of the time series image capturing of described classification classification image and different classes of correspondence:
The time changing curve of all pixels in each classification of time series image capturing according to each classification；
Each time point of all pixels in each classification is averaged and obtains the value of each time point in each classification；
The time series variation curve of each classification is obtained according to the value of each time point in described each classification.
5. processing method according to claim 1, wherein, also includes:
Obtain the pre-set categories in described classification image according to operational order, time series variation curve corresponding for described pre-set categories is highlighted；
Obtain the Preset Time sequence variation curve in the time series variation curve of different classes of correspondence according to operational order, classification corresponding for described Preset Time sequence variation curve is highlighted in described classification image.
6. processing method according to claim 5, wherein, described classification image includes corresponding positional information or identification information with described time series variation curve.
7. processing method according to claim 6, wherein, is highlighted time series variation curve corresponding for described pre-set categories and includes:
Obtain the positional information of pre-set categories；
The time series variation curve corresponding with described positional information is obtained according to described positional information；
The time series variation curve corresponding with described positional information is highlighted.
8. processing method according to claim 6, wherein, is highlighted time series variation curve corresponding for described pre-set categories and includes:
Obtain the identification information of pre-set categories；
The time series variation curve corresponding with described identification information is obtained according to described identification information；
The time series variation curve corresponding with described identification information is highlighted.
9. processing method according to claim 6, wherein, is highlighted classification corresponding for described Preset Time sequence variation curve in described classification image and includes:
Obtain the positional information of Preset Time sequence variation curve；
The classification corresponding with described positional information is obtained according to described positional information；
The classification corresponding with described positional information is highlighted by described classification image.
10. processing method according to claim 6, wherein, is highlighted classification corresponding for described Preset Time sequence variation curve in described classification image and includes:
Obtain the identification information of Preset Time sequence variation curve；
The classification corresponding with described identification information is obtained according to described identification information；
The classification corresponding with described identification information is highlighted by described classification image.
11. processing method according to claim 1, wherein, also include:
By described classification image store in the first data base；
The time series variation curve of different classes of correspondence is stored in the second data base.
12. processing method according to claim 1, wherein, the time series variation curve of described different classes of correspondence graphically shows.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610119032.2A CN105809189A (en) | 2016-03-02 | 2016-03-02 | Time series image processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610119032.2A CN105809189A (en) | 2016-03-02 | 2016-03-02 | Time series image processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105809189A true CN105809189A (en) | 2016-07-27 |
Family
ID=56466562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610119032.2A Pending CN105809189A (en) | 2016-03-02 | 2016-03-02 | Time series image processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105809189A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108460789A (en) * | 2018-03-19 | 2018-08-28 | 国家基础地理信息中心 | A kind of artificial earth's surface timing variations on-line detecting system and method |
| CN109145144A (en) * | 2018-08-24 | 2019-01-04 | 贵州宽凳智云科技有限公司北京分公司 | The matching process of position and picture when high-precision road data acquires |
| CN110232331A (en) * | 2019-05-23 | 2019-09-13 | 深圳大学 | A kind of method and system of online face cluster |
| CN114792116A (en) * | 2022-05-26 | 2022-07-26 | 中国科学院东北地理与农业生态研究所 | Time series deep convolution network crop remote sensing classification method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239884A (en) * | 2014-08-29 | 2014-12-24 | 中国科学院遥感与数字地球研究所 | Abnormal submerging area detection method based on remote sensing vegetation index time sequence |
-
2016
- 2016-03-02 CN CN201610119032.2A patent/CN105809189A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239884A (en) * | 2014-08-29 | 2014-12-24 | 中国科学院遥感与数字地球研究所 | Abnormal submerging area detection method based on remote sensing vegetation index time sequence |
Non-Patent Citations (3)
| Title |
|---|
| XIN ZHANG 等: "WEBGIS based CropWatch online agriculture monitoring system", 《HTTPS://AGU.CONFEX.COM/AGU/FM15/WEBPROGRAM/PAPER77953.HTML》 * |
| 吴炳方: "中国农情遥感速报系统", 《遥感学报》 * |
| 邹文涛 等: "农作物长势综合监测——以印度为例", 《遥感学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108460789A (en) * | 2018-03-19 | 2018-08-28 | 国家基础地理信息中心 | A kind of artificial earth's surface timing variations on-line detecting system and method |
| CN108460789B (en) * | 2018-03-19 | 2020-05-26 | 国家基础地理信息中心 | Artificial earth surface time sequence change on-line detection system and method |
| CN109145144A (en) * | 2018-08-24 | 2019-01-04 | 贵州宽凳智云科技有限公司北京分公司 | The matching process of position and picture when high-precision road data acquires |
| CN109145144B (en) * | 2018-08-24 | 2021-07-30 | 贵州宽凳智云科技有限公司北京分公司 | Method for matching position and picture during high-precision road data acquisition |
| CN110232331A (en) * | 2019-05-23 | 2019-09-13 | 深圳大学 | A kind of method and system of online face cluster |
| CN110232331B (en) * | 2019-05-23 | 2022-09-27 | 深圳大学 | Online face clustering method and system |
| CN114792116A (en) * | 2022-05-26 | 2022-07-26 | 中国科学院东北地理与农业生态研究所 | Time series deep convolution network crop remote sensing classification method |
| CN114792116B (en) * | 2022-05-26 | 2024-05-03 | 中国科学院东北地理与农业生态研究所 | Remote sensing classification method for crops in time sequence deep convolution network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gašparović et al. | An automatic method for weed mapping in oat fields based on UAV imagery | |
| Chen et al. | Airborne LiDAR remote sensing for individual tree forest inventory using trunk detection-aided mean shift clustering techniques | |
| Meneguzzo et al. | Mapping trees outside forests using high-resolution aerial imagery: a comparison of pixel-and object-based classification approaches | |
| CN111630551A (en) | Crop type classification in images | |
| CN105809189A (en) | Time series image processing method | |
| JP6273473B2 (en) | Forest information management device | |
| WO2009040063A1 (en) | Apparatus and method for updating map data | |
| Röder et al. | Application of optical unmanned aerial vehicle-based imagery for the inventory of natural regeneration and standing deadwood in post-disturbed spruce forests | |
| DeWitt et al. | Creating high-resolution bare-earth digital elevation models (DEMs) from stereo imagery in an area of densely vegetated deciduous forest using combinations of procedures designed for lidar point cloud filtering | |
| Gomes et al. | Detection of tree crowns in very high spatial resolution images | |
| CN104902438A (en) | Statistical method and system for analyzing passenger flow characteristic information on the basis of mobile communication terminal | |
| CN113240009A (en) | Point cloud data labeling method and device, storage medium and electronic equipment | |
| Grigillo et al. | Automated building extraction from IKONOS images in suburban areas | |
| Vatandaşlar et al. | Extraction of forest inventory parameters using handheld mobile laser scanning: A case study from Trabzon, Turkey | |
| Lehmann et al. | Temporal and spatial changes of land use in rare metal mining areas of Rwanda | |
| Kalantar et al. | Smart counting–oil palm tree inventory with UAV | |
| Gholoobi et al. | Using object-based hierarchical classification to extract land use land cover classes from high-resolution satellite imagery in a complex urban area | |
| Ayrey et al. | Ecologically-based metrics for assessing structure in developing area-based, enhanced forest inventories from LiDAR | |
| Liu et al. | A framework of road extraction from airborne lidar data and aerial imagery | |
| Belcore et al. | Riparian ecosystems mapping at fine scale: a density approach based on multi‐temporal UAV photogrammetric point clouds | |
| CN105740901B (en) | Mutative scale object-oriented Classification in Remote Sensing Image antidote based on ontology | |
| Prerna et al. | Evaluation of LiDAR and image segmentation based classification techniques for automatic building footprint extraction for a segment of Atlantic County, New Jersey | |
| WO2023116359A1 (en) | Method, apparatus and system for classifying green, blue and gray infrastructures, and medium | |
| CN113723833B (en) | Method, system, terminal equipment and storage medium for evaluating quality of forestation actual results | |
| Hirschmugl | Derivation of forest parameters from UltracamD data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160727 |
|
| RJ01 | Rejection of invention patent application after publication |