CN103279954A - Remote-sensing image change detecting method based on land utilization database - Google Patents
Remote-sensing image change detecting method based on land utilization database Download PDFInfo
- Publication number
- CN103279954A CN103279954A CN2013101893447A CN201310189344A CN103279954A CN 103279954 A CN103279954 A CN 103279954A CN 2013101893447 A CN2013101893447 A CN 2013101893447A CN 201310189344 A CN201310189344 A CN 201310189344A CN 103279954 A CN103279954 A CN 103279954A
- Authority
- CN
- China
- Prior art keywords
- image
- land
- frequency
- variation
- characteristic
- 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
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the field of remote-sensing image change detection and provides a remote-sensing image change detecting method based on a land utilization database in order to adapt to the requirement of national land change survey and quickly extract newly added construction land through indoor work. The technical scheme is that the land utilization database is utilized, the boundary of the land utilization database serves as an object cut-oriented cutting result, and the cutting precision is improved; a characteristic value of land type pattern spot data serves as a sample statistical empirical value of various types of land features, and a threshold value is provided for a characteristic image difference value; a detecting method based on a pixel and oriented to a specific target is combined to provide a method for extracting a green plant covered region from a real colored image, the land type attribute is helped to be judged, and the precision of change detection is improved; and according to the energy distribution characteristic of an image in a frequency domain, a method for self-adaptively and precisely extracting image detail characteristics is provided, and the construction land is quickly extracted according to the detail characteristics.
Description
Technical field
The invention belongs to remote sensing image and change detection range, particularly based on the remote sensing image change detecting method in land use data storehouse.
Background technology
According to country's " land investigation management " regulation, country carries out soil change investigation every year.Soil change investigation is important national conditions and strength investigation, is effective foundation of implementing land resources planning, management, protection and rationally utilizing, is the basis of establishment the national economic and social development planning, relevant ad hoc planning.The main stream approach of the investigation of soil change at present, the artificial relatively changing unit of two width of cloth images of industry is sketched the contours it again in being still, and last field investigation is mended and is painted.The characteristics huge at artificial variation testing amount, that the work period is long are applied in the soil change automatic change detection techniques very urgent.
Summary of the invention
At the big defective of manual detection region of variation workload in the change investigation of territory department soil, the object of the present invention is to provide a kind of change detecting method fast and accurately based on the land use data storehouse, improve the variation detection efficiency of land investigation change.
For achieving the above object, variation of the present invention detects, and based on the land use data storehouse, will detect based on the variation of pixel and combines based on the variation detection of specific objective, draws final variation testing result.It is characterized in that, the image pretreatment module is provided, comprise image denoising and relative radiant correction; Change detection module based on pixel is provided, and the phase image extracts eigenwert during to front and back, and utilizes differential technique to change and detect; Change detection module based on the part specific objective is provided, has specifically comprised the extraction of atural objects such as vegetation, cloud; Variation accuracy of detection inspection module is provided.
The image pretreatment module comprises image denoising and relative radiant correction.The both is instrument with the wavelet transformation, handles image again in wavelet field.Wherein, the image denoising adopts wavelet field to decompose image, then the image HFS is adopted the soft-threshold denoising.Radiant correction to the coupling of two width of cloth image low frequency part histograms after the wavelet decomposition, makes the intensity profile of two width of cloth images roughly the same relatively.
Change detection module based on pixel.The present invention adopts the characteristic image differential technique, few at the shared number of pixels of single solitary building in the meter level resolution remote sense image, buildings and surrounding environment luminance contrast are big, characteristics such as the bigger and brightness of buildings intrinsic brightness is even, adopt Fourier transform to extract the high-frequency characteristic of image, then most buildingss all are extracted out, again to two width of cloth Fourier characteristic image differences, the difference image is set suitable empirical value again, draw preliminary variation testing result, utilize image morphology to handle at last, especially by calculating the profile that changes image, utilize profile can calculate the characteristics of segment area fast, deletion finally can draw the variation testing result fast less than the variation segment of the area of pictural surface on the land investigation.
Based on the change detection module of specific objective, can effectively detect the high cloud cluster in the image, and can accurately detect the territory, vegetation-covered area.Be the hsv color space with the RGB color space conversion, utilize H component setting threshold, and with certain image leak strategy of filling up, can effectively detect high cloud cluster zone, determine to change the inactive area that detects.With H component and S component difference, the gray scale of image vegetation and other ground classes differs greatly after the difference.Therefore, by setting zero threshold value, just can be in visible-range territory, very accurate extraction vegetation-covered area.The variation testing result of this module provides strong circumstantial evidence for final variation testing result.
The accuracy test module, according to the feature of the variation segment that detects, relatively itself and the feature of various places class calculate the probability that segment belongs to the various places class, and go out to detect the degree of confidence of this variation segment by these probability calculations.
Description of drawings
Fig. 1 is the flow process that whole variation detects.
Embodiment
The remote sensing image change detecting method based on the land use data storehouse that the present invention proposes is characterized in that, comprises the image pretreatment module, specifically comprises image denoising and relative radiant correction.
Step 1 is utilized wavelet decomposition technology and soft-threshold treatment technology, and by extracting the image high-frequency information, in the HFS denoising, the wavelet decomposition denoising comprises following step,
Step 1.1 raw video utilizes Lifting Wavelet to decompose;
Step 1.2 is asked threshold value according to the decomposition number of plies of setting to the every layer image after decomposing;
Step 1.3 is according to the threshold value of every layer image, to every layer image difference threshold value;
The image small echo recovers after the step 1.4 pair threshold value, obtains final denoising image.
Step 2 is utilized wavelet decomposition technology and histogram matching technique, and by the relative histogram of image low frequency part is mated, small echo recovers again.To image histogram coupling, be divided into following step,
Step 2.1 is asked for the histogram of two width of cloth images respectively;
Step 2.2 is asked the accumulation histogram probability density of two width of cloth images;
Step 2.3 to each gray scale cumulative probability value of being mated image, is searched the gray-scale value of cumulative probability density and its difference minimum in the image to be matched to be standard by the coupling image, and changes its gray scale for being mated the gray scale of image.
Step 3 pair pretreated image carries out cloud detection and territory, vegetation-covered area and extracts, and to detecting the zone of cloud, changes when detecting being thought of as inactive area, can't detect; To detecting the zone that is covered by vegetation, determine the ground generic attribute of corresponding segment as standard.Concrete performing step is as follows,
Step 3.1 is converted to the HSV space with the image rgb space, and concrete conversion formula is
Step 3.2 is obtained the difference image with H component and the S component difference in HSV space;
Step 3.3 pair difference image threshold value, threshold value is set at 0 and gets final product, and is the territory, vegetation-covered area greater than the pixel of threshold value;
Step 3.4 pair H component setting threshold, and utilize morphology medium filtering, opening and closing operation, fill up methods such as leak, detect the high cloud overlay area.
Step 4 pair pretreated image carries out detecting based on the variation of pixel.Variation based on pixel detects, and comprises the characteristic information that extracts raw video, extracts change information by the characteristic information difference again.Specifically may further comprise the steps,
Step 4.1 pair raw video Fourier transform adopts Butterworth high-pass filtering function again, extracts the high-frequency information of image.The filtering cutoff frequency of high-pass filtering is determined according to the energy percentage of image in the frequency domain.After being frequency field according to image by space field transformation, the image energy by around the middle mind-set, frequency is transformed to high frequency, image medium and low frequency information gradually by low frequency and accounts for most characteristics, be initial point with frequency domain image center point, calculate the image energy percentage in the certain radius scope, determine that the radius of a circle when energy reaches gross energy 92% corresponds to cutoff frequency;
The high-frequency information difference of step 4.2 pair extraction is chosen appropriate threshold again, draws preliminary change detection result.Choosing by land use data storehouse cutting image of threshold value calculated gray scale and other characteristic statistics information of various ground class image blocks again and determined;
Step 4.3 in conjunction with empirical value, is deleted undesirable surveyed area according to the requirement of soil change investigation.Undesirable surveyed area comprises: being changed to the zone of green vegetation, being changed to the zone in waters, is the zone of buildings before changing.
Step 5 is according to the feature of the variation segment that detects, and relatively itself and the feature of various places class calculate the probability that segment belongs to the various places class, and go out to detect the degree of confidence of this variation segment by these probability calculations.
Claims (4)
1. one kind with the land use data storehouse with based on the meter level resolution true color image change detecting method that pixel characteristic image differential technique combines, and it is characterized in that comprising following steps:
Step 1 according to the land use data storehouse, is calculated the differently feature empirical value of class;
Step 2 is utilized Fourier transform, and the characteristic image of phase image is characterized in that in the time of before and after calculating, and every width of cloth image is carried out following steps,
Step 2.1 image Fourier direct transform, with video conversion in frequency domain;
Step 2.2 to the image High frequency filter, is utilized the Butterworth Hi-pass filter in frequency domain, extract the high-frequency characteristic of image, again to the image inverse Fourier transform in spatial domain;
Step 3, the phase image extracts green overlay area during to the back, is defined as non-region of variation;
Step 4 to two width of cloth image differences, according to the various places class statistics empirical value that calculates in the step 1, excludes three class region of variation again: variation is preceding to be construction area, change the back is to be the waters after territory, vegetation-covered area, the variation.
2. a self-adaptation cutoff frequency obtains the image high-frequency information, to generate the method for minutia image, it is characterized in that: common High Pass Filter Cutoff Frequency all wants artificial empirical value given, and the frequency cutoff frequency among the present invention is to ask for according to the information feature self-adaptation of different images itself, specifically may further comprise the steps:
Step 2.1 utilizes Fourier transform that the spatial domain image is transformed in the frequency field;
Step 2.2 under the ever-increasing situation, is initial point with the image center at the frequency field image from the center to the marginal frequency, also calculates the energy percentage that the circle self-energy accounts for the view picture image to the external diffusion radius, and the radius that reaches at 92% o'clock is the high-pass filtering cutoff frequency.
3. one kind is extracted the method for green mulched ground to true color image, and its feature comprises following step:
Step 3.1, with true color image from the RGB color notation conversion space to the hsv color space;
Step 3.2, with the H component in the HSV space and S component difference, setting threshold is 0, is vegetation greater than the pixel areal coverage of threshold value, otherwise is other ground class.
4. one kind is utilized the land use data storehouse to calculate the differently method of category feature empirical value, it is characterized in that comprising following steps:
Step 4.1 is utilized the remote sensing image of land-use map spot database and contemporaneity, utilizes figure spot border cutting image according to the plat spot;
Step 4.2, looking like with each figure fleck is unit, calculates all figure spot image characteristic values, comprises gray average, variance etc.;
Step 4.3 according to the figure spot ground generic attribute in the land use data storehouse, is calculated gray average, variance statistical value and other eigenwerts of every kind of ground class.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101893447A CN103279954A (en) | 2013-05-21 | 2013-05-21 | Remote-sensing image change detecting method based on land utilization database |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101893447A CN103279954A (en) | 2013-05-21 | 2013-05-21 | Remote-sensing image change detecting method based on land utilization database |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103279954A true CN103279954A (en) | 2013-09-04 |
Family
ID=49062461
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013101893447A Pending CN103279954A (en) | 2013-05-21 | 2013-05-21 | Remote-sensing image change detecting method based on land utilization database |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103279954A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106059492A (en) * | 2016-05-05 | 2016-10-26 | 江苏方天电力技术有限公司 | Photovoltaic assembly shadow fault type determination method based on power prediction |
| CN106156756A (en) * | 2016-07-28 | 2016-11-23 | 广州地理研究所 | The Method of fast estimating of construction land efficiency spatial distribution |
| CN107341795A (en) * | 2017-06-30 | 2017-11-10 | 武汉大学 | A kind of high spatial resolution remote sense image method for detecting automatic variation of Knowledge driving |
| CN110599479A (en) * | 2019-09-16 | 2019-12-20 | 北京航天宏图信息技术股份有限公司 | Monitoring area change detection method and device and storage medium |
| CN112967286A (en) * | 2021-05-19 | 2021-06-15 | 航天宏图信息技术股份有限公司 | Method and device for detecting newly added construction land |
| CN113255808A (en) * | 2021-06-03 | 2021-08-13 | 中国科学院地理科学与资源研究所 | Long-time-sequence territorial space regional functional structure change detection method based on big data |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101694718A (en) * | 2009-10-13 | 2010-04-14 | 西安电子科技大学 | Method for detecting remote sensing image change based on interest areas |
| CN102169584A (en) * | 2011-05-28 | 2011-08-31 | 西安电子科技大学 | Remote sensing image change detection method based on watershed and treelet algorithms |
-
2013
- 2013-05-21 CN CN2013101893447A patent/CN103279954A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101694718A (en) * | 2009-10-13 | 2010-04-14 | 西安电子科技大学 | Method for detecting remote sensing image change based on interest areas |
| CN102169584A (en) * | 2011-05-28 | 2011-08-31 | 西安电子科技大学 | Remote sensing image change detection method based on watershed and treelet algorithms |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106059492A (en) * | 2016-05-05 | 2016-10-26 | 江苏方天电力技术有限公司 | Photovoltaic assembly shadow fault type determination method based on power prediction |
| CN106059492B (en) * | 2016-05-05 | 2018-03-06 | 江苏方天电力技术有限公司 | Photovoltaic module shade fault type judges method based on power prediction |
| CN106156756A (en) * | 2016-07-28 | 2016-11-23 | 广州地理研究所 | The Method of fast estimating of construction land efficiency spatial distribution |
| CN107341795A (en) * | 2017-06-30 | 2017-11-10 | 武汉大学 | A kind of high spatial resolution remote sense image method for detecting automatic variation of Knowledge driving |
| CN107341795B (en) * | 2017-06-30 | 2020-03-10 | 武汉大学 | Knowledge-driven high-spatial-resolution remote sensing image automatic change detection method |
| CN110599479A (en) * | 2019-09-16 | 2019-12-20 | 北京航天宏图信息技术股份有限公司 | Monitoring area change detection method and device and storage medium |
| CN112967286A (en) * | 2021-05-19 | 2021-06-15 | 航天宏图信息技术股份有限公司 | Method and device for detecting newly added construction land |
| CN113255808A (en) * | 2021-06-03 | 2021-08-13 | 中国科学院地理科学与资源研究所 | Long-time-sequence territorial space regional functional structure change detection method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108846832B (en) | Multi-temporal remote sensing image and GIS data based change detection method and system | |
| CN103824077B (en) | Urban impervious layer rate information extraction method based on multi-source remote sensing data | |
| CN103578110B (en) | Multiband high-resolution remote sensing image dividing method based on gray level co-occurrence matrixes | |
| CN104881865B (en) | Forest pest and disease monitoring method for early warning and its system based on unmanned plane graphical analysis | |
| CN103279954A (en) | Remote-sensing image change detecting method based on land utilization database | |
| CN112164062A (en) | Wasteland information extraction method and device based on remote sensing time sequence analysis | |
| CN104182754B (en) | A kind of rural residential area information extracting method based on high-resolution remote sensing image | |
| Zhou et al. | Individual tree parameters estimation for plantation forests based on UAV oblique photography | |
| CN102819740B (en) | A kind of Single Infrared Image Frame Dim targets detection and localization method | |
| CN101937079A (en) | Remote sensing image variation detection method based on region similarity | |
| CN111339948B (en) | Automatic identification method for newly-added buildings of high-resolution remote sensing images | |
| CN102842044B (en) | Method for detecting variation of remote-sensing image of high-resolution visible light | |
| CN102855487A (en) | Method for automatically extracting newly added construction land change image spot of high-resolution remote sensing image | |
| CN104361582A (en) | Method of detecting flood disaster changes through object-level high-resolution SAR (synthetic aperture radar) images | |
| CN104463836A (en) | City green space remote-sensing measuring method based on moving windows | |
| CN109377480A (en) | Arable land use change detection method based on deep learning | |
| CN104850853B (en) | City rapid extracting method and device | |
| CN108052876A (en) | Regional development appraisal procedure and device based on image identification | |
| CN109671038B (en) | Relative radiation correction method based on pseudo-invariant feature point classification layering | |
| Zhang et al. | Methods for automatic identification and extraction of terraces from high spatial resolution satellite data (China-GF-1) | |
| CN102567726B (en) | Technology for automatically extracting floating ice in polar ice sheet edge regions | |
| CN104573662B (en) | A kind of cloud sentences method and system | |
| CN104517286A (en) | SAR (synthetic aperture radar) image registration based on self-adaption threshold segmentation and combination optimization | |
| Xu et al. | The comparative study of three methods of remote sensing image change detection | |
| CN107067384B (en) | A kind of loess plateau terraced fields extracting method based on ray method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130904 |