CN101694718A - Method for detecting remote sensing image change based on interest areas - Google Patents
Method for detecting remote sensing image change based on interest areas Download PDFInfo
- Publication number
- CN101694718A CN101694718A CN200910024294A CN200910024294A CN101694718A CN 101694718 A CN101694718 A CN 101694718A CN 200910024294 A CN200910024294 A CN 200910024294A CN 200910024294 A CN200910024294 A CN 200910024294A CN 101694718 A CN101694718 A CN 101694718A
- Authority
- CN
- China
- Prior art keywords
- edge
- differential chart
- interest
- class
- pixel
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a method for detecting remote sensing image change based on interest areas, which belongs to the fields of analyzing and processing remote sensing images. The invention mainly solves the problems of pseudo-change information existing in the methods for detecting remote sensing image change, the implementation process comprises: firstly, extracting the outline of a differential chart, secondly, defining the outline map of a closed interest area based on the threshold value outline map chained method, and obtaining the interest area after expanding the outline, thirdly, obtaining a new differential chart through updating the differential chart according to the gray level, the space position and the sort feature of the interest area and a non-interest area, fourthly, dividing the interest area in the new differential chart, and obtaining a change result chart. The method has the advantages of accurate location of the interest area, noise and registration error robust, less pseudo-change information and high change detection, and can be used in the fields of disaster surveillance, land utilization and agriculture investigation.
Description
Technical field
The invention belongs to technical field of image processing, specifically a kind of method of the Remote Sensing Imagery Change Detection based on area-of-interest is applicable to remote Sensing Image Analysis and processing.
Background technology
The change-detection of remote sensing images is meant by analyzing at two width of cloth or several remote sensing images of different time from areal, detects the change information that the atural object of this area takes place in time.Development along with remote sensing and infotech, change-detection has been widely used in the dynamic monitoring as the forest reserves, the variation monitoring that the soil covers, utilizes, the agricultural resource investigation, urban planning layout, environmental monitoring and analysis, disaster assessment, many fields such as dynamic surveillance of strategic objectives such as road, bridge, airport in geographic data updates and the military surveillance.
In existing change detecting method, the method for analyzing based on difference image is different from additive method because directly simple, can not be loyal to raw data owing to method itself to detecting preceding change information change, has guaranteed change-detection result's reliability.Yet owing to causing not alternate simultaneously gradation of image value, factors such as the illumination under Various Seasonal and the situation, radiation do not have deviation whole or part between the remote sensing images of phase simultaneously, therefore the disparity map that computing obtains to gray-scale value is simply carried out Threshold Segmentation, has a lot of pseudo-change informations among the resulting change-detection result.
For solving the problem of above-mentioned existence, some scholars have proposed the improvement to the disparity map dividing method:
One, scholar such as Bruzzone proposed to ask threshold value to the disparity map classification and adopt markov random file to analyze the spatial neighborhood relation of disparity map based on Bayes minimum error probability in article " Automatic Analysis of the Difference Image forUnsupervised Change Detection " in 2000, obtain change-detection figure as a result, to reach the effect of removing pseudo-change information.This method is the method for typically analyzing based on difference image, can access change-detection result preferably, yet the initial markers mistake of disparity map can cause last change-detection erroneous results, thereby influence last change-detection precision as a result.
Two, scholar such as the Fang Shenghui edge that in article " based on the change detecting method research of edge feature ", goes out region of variation according to the edge variation feature and the grey scale change feature extraction of 2 o'clock phase remote sensing images, and with region of variation mark in former figure.This method shortcoming is the much disconnected edge of the marginal existence of the region of variation that obtains, and detects the edge of a lot of false region of variation.
Summary of the invention
The objective of the invention is to overcome the deficiency of above-mentioned existing Remote Sensing Imagery Change Detection technology, a kind of method for detecting change of remote sensing image based on area-of-interest has been proposed, accurately to detect closed region of variation outline map, reduced pseudo-change information, improved the change-detection precision.
For achieving the above object, detection method of the present invention comprises the steps:
1) to the input two width of cloth not simultaneously the phase remote sensing images carry out the medium filtering denoising respectively;
The image corresponding pixel points gray scale difference value of phase calculates during 2) to filtered two, obtains a width of cloth differential chart;
3) the Canny edge of calculated difference figure, and calculate the characteristics of mean of each marginal point, obtain a breadths edge average figure;
4) with the fuzzy C Mean Method edge average figure is divided into non-certainly variation class, unmarked class and changes class certainly, and calculate pixel value maximum in the unmarked class and change the mean value of the minimum pixel value in the class as high threshold certainly, calculate the low threshold value of mean value conduct of affirming pixel value maximum in the non-variation class and the minimum pixel value in the unmarked class, according to high threshold and low threshold value edge average figure is cut apart respectively, obtained the low threshold value outline map of a panel height threshold value outline map and a width of cloth;
5) according to dual threshold boundary chain connection, the disconnected edge in the link high threshold outline map, obtaining the edge is the boundary chain map interlinking of enclosed region;
6) enclosed region in the boundary chain map interlinking of gained is inwardly outwards respectively expanded two pixels respectively along the normal orientation at its edge, enclosed region after expanding is carried out the zone fills, and the enclosed region after will filling is as area-of-interest, with other zones as non-area-of-interest;
7) with the minimal error rate threshold method pixel in the differential chart is divided into and changes class and non-variation class, differential chart is upgraded, obtain new differential chart according to classification, locus and the neighborhood relationships of the pixel of area-of-interest in the differential chart and non-area-of-interest;
8) calculate the maximum between-cluster variance threshold value of new differential chart and its area-of-interest is divided into changes class and non-variation class, the figure as a result of change-detection to the end.
The present invention has the following advantages compared with prior art:
(1) the present invention is owing to adopt the dual threshold edge to connect and the recurrence track algorithm, the edge line that the high threshold outline map is interrupted connects into enclosed region, and the edge of the enclosed region that obtains inwardly outwards expanded, thereby guaranteed to obtain area-of-interest accurately, avoided the influence of 2 o'clock phase images registration errors to the change-detection result.
(2) the present invention is because according to the pixel value in preliminary classification classification, locus and the neighborhood relationships renewal differential chart of area-of-interest and non-area-of-interest pixel, it is more accurate to make that the differential chart threshold value obtains.
(3) the present invention has avoided the view picture differential chart is cut apart the pseudo-change information that is caused owing to only area-of-interest is cut apart, and has improved the accuracy of change-detection.
Description of drawings
Fig. 1 is realization flow figure of the present invention;
Phase remotely-sensed data image when Fig. 2 is the present invention two;
Fig. 3 is the differential chart of 2 o'clock phasors behind the medium filtering of the present invention;
Fig. 4 is the Canny outline map of differential chart of the present invention;
Fig. 5 is high threshold outline map of the present invention and low threshold value outline map;
Fig. 6 is an edge of the present invention connection layout;
Fig. 7 is an area-of-interest of the present invention;
Fig. 8 is the new differential chart of the present invention;
Fig. 9 is change-detection figure and the change-detection reference diagram as a result as a result of the present invention and existing method of contrast.
Embodiment
With reference to Fig. 1, enforcement of the present invention is as follows:
Step 1 is imported the not remote sensing images of phase simultaneously of two width of cloth, and shown in Fig. 2 (a) and Fig. 2 (b), it is 3 * 3 medium filterings that two width of cloth are not carried out the filter window size respectively simultaneously mutually, obtains not phasor simultaneously of filtered two width of cloth.
Step 2 is calculated two width of cloth gray scale difference value between the corresponding pixel points of phasor simultaneously not behind the medium filtering, obtains a width of cloth differential chart, as shown in Figure 3.
Step 3, the Canny edge of calculated difference figure, and calculate the characteristics of mean of each marginal point, and obtaining a breadths edge average figure, concrete steps are as follows:
(3a) adopt the Canny boundary operator, obtain the Canny outline map of a width of cloth differential chart, as shown in Figure 4 the differential chart computing.
Be the center (3b) with each marginal point in the Canny outline map of differential chart, calculating in 3 * 7 moving window 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° of line segment lengths that marginal point is formed in totally eight directions; the line segment of getting edge line segment length maximum is as the edge line in this window; the pairing direction of this edge line is as current marginal point place edge direction; and the marginal point on the non-edge line in the moving window removed; and be labeled as non-marginal point at the correspondence position of the Canny of differential chart outline map, thus obtain new Canny outline map;
(3c) calculate the pixel average of new each marginal point of Canny outline map on 3 * 7 moving window inward flange line both sides of differential chart correspondence position respectively, average that will be bigger is as the characteristics of mean of this marginal point, thereby obtains a breadths edge average figure.
Step 4 is determined the high threshold of edge average figure, low threshold value according to the fuzzy C Mean Method, and with high threshold, low threshold value edge average figure is cut apart respectively and obtained a panel height threshold value outline map and a width of cloth hangs down the threshold value outline map.
Being implemented as follows of this step:
(4a) edge average figure is divided into non-certainly variation class, unmarked class and changes class certainly with the fuzzy C Mean Method;
(4b) with pixel value maximum in the unmarked class among the edge average figure and the mean value that changes the minimum pixel value in the class certainly as high threshold, will be among the edge average figure certainly in the non-variation class mean value of maximum pixel value and the minimum pixel value in the unmarked class as low threshold value;
(4c) according to high threshold and low threshold value edge average figure is cut apart respectively, if pixel value is greater than high threshold among the edge average figure, the marginal point that then keeps relevant position in the Canny outline map of differential chart, obtain a panel height threshold value outline map, shown in Fig. 5 (a),, then keep the marginal point of relevant position in the Canny outline map of differential chart if pixel value is greater than low threshold value among the edge average figure, obtain the low threshold value outline map of a width of cloth, shown in Fig. 5 (b).
Step 5 according to dual threshold edge connection method, connects into enclosed region with the disconnected edge in the high threshold outline map, and obtaining the edge is the edge connection layout of enclosed region.
(5a) from low threshold value outline map, the high threshold outline map is deducted, obtains the edge differential chart, and with the high threshold outline map as the edge connection layout;
(5b) the disconnected marginal point in the search edge connection layout in the edge differential chart, if in the edge differential chart, there is marginal point to be connected in eight neighborhoods of these marginal points, in the edge connection layout, be marginal point then with the position mark of this neighborhood with discontinuity edge line in the high threshold outline map;
(5c) according to recurrence track algorithm continuous repeating step (5b) in the edge differential chart, edges all in obtaining the edge connection layout are enclosed region, as shown in Figure 6.
Step 6, each marginal point to the enclosed region of edge connection layout, inwardly outwards respectively expand two pixels respectively along its edge normal orientation, enclosed region after expanding is carried out the zone fills, and the enclosed region after will filling is as area-of-interest, as shown in Figure 7, in Fig. 7, be area-of-interest with white marking, density bullet is non-area-of-interest.
Step 7 is classified to pixel in the differential chart, and according to classification, locus and the neighborhood relationships of the pixel of area-of-interest and non-area-of-interest differential chart is upgraded, and obtains new differential chart.
Being implemented as follows of this step:
(7a) adopt minimal error rate threshold method calculated difference figure threshold value T, if gray values of pixel points is greater than threshold value T in the differential chart, then with this pixel as changing class, otherwise as non-variation class;
(7b) whether the locus of judging pixel in the differential chart in area-of-interest, if current pixel point not in area-of-interest, then will be that intermediate value in 5 * 5 sliding windows at center is as the difference of current pixel point with the current pixel point; Otherwise, current pixel point in area-of-interest, execution in step (7c);
(7c) whether the classification of judging pixel in the differential chart belongs to the variation class, if belong to the variation class, then will be that maximal value in 5 * 5 sliding windows at center is as the difference of current point with the current pixel point; Otherwise this pixel belongs to non-variation class, keeps the initial value in the differential chart, thereby obtain new differential chart, as shown in Figure 8, the gray-scale value contrast of the gray-scale value of region of variation and non-region of variation strengthens in the new differential chart that obtains, and makes region of variation highlight in new differential chart.
Step 8 is calculated the maximum between-cluster variance threshold value of new differential chart, according to this threshold value with in the area-of-interest in the new differential chart greater than the pixel of threshold value as changing class, otherwise as non-variation class, result of variations to the end, shown in Fig. 9 (a).
Effect of the present invention can further specify by following experimental result and analysis:
1. experimental data
The Landsat remote sensing image data collection of phase when experimental data of the present invention is three groups of single bands two.Phasor is that the Landsat-5TM image obtains data in August, 1994 and in September, 1994 in the western part on Elba island respectively during first group of data two, the time mutually 1 image such as Fig. 2 (a), the time mutually 2 images shown in Fig. 2 (b), the image size is 325 * 414,256 gray levels, Fig. 9 (e) is the variation reference diagram of Fig. 2 (a) and 2 (b), and white portion is represented the zone that changes among the figure.Second group of data is 2 o'clock phase Landsat7ETM+4 wave band remote sensing images in Mexico countryside, and size is 512 * 512,256 gray levels.The 3rd group data set is 2 o'clock phase Landsat5TM+5 band images on certain island of Italy, and size is 412 * 300,256 gray levels.
2. contrast experiment and experimental evaluation index
Method of contrast is that method that scholar such as Bruzzone proposed in article " Automatic Analysis of the DifferenceImage for Unsupervised Change Detection " in 2000 is carried out the difference computing with 2 o'clock phase remote sensing images and obtained differential chart, and differential chart is carried out Threshold Segmentation obtain change-detection figure as a result, be called differential technique, on the basis of differential technique, adopt markov random file to consider that the spatial neighborhood relation of differential chart obtains the change-detection result, is called the markov random file method then.
The present invention designs two experiments and verifies validity of the present invention.In order to verify validity of the present invention under the prerequisite that adopts same threshold, first experiment is to adopt the maximum between-cluster variance threshold method that the differential chart of the present invention and differential technique is cut apart the change-detection result who obtains to compare, verify of the influence of the detected area-of-interest of the present invention to the change-detection result, change-detection wherein of the present invention is figure as a result, shown in Fig. 9 (a), the change-detection that the differential technique method obtains as a result figure shown in Fig. 9 (b).Second experiment be the present invention adopt change-detection that markov random file considers spatial relationship as a result the change-detection that obtains of figure and markov random file method as a result figure compare, verify that the present invention considers the validity under the prerequisite of spatial relationship, change-detection wherein of the present invention is figure as a result, shown in Fig. 9 (c), the change-detection that the markov random file method obtains is figure as a result, shown in Fig. 9 (d).
Analysis in the experiment to change-detection result's amount of carrying out and matter.The evaluation index of amount comprises false-alarm number, omission number and total wrong number, the evaluation of matter be with change-detection as a result Fig. 9 (a), Fig. 9 (b), Fig. 9 (c), Fig. 9 (d) with carry out the subjective vision contrast with reference to figure 9 (e).Table 1 is the evaluation index of the inventive method and differential technique method in first experiment, and table 2 is evaluation indexes of the inventive method and markov random file method in second experiment.
3. experimental result and analysis
In first experiment, the change-detection result that the present invention obtains is shown in Fig. 9 (a), and the change-detection result that the differential technique method obtains is shown in Fig. 9 (b), and reference diagram is shown in Fig. 9 (e).From Fig. 9 (a) and Fig. 9 (e) more as can be seen, among Fig. 9 (a) pseudo-change information seldom, and keep on the details also fine, approach most with reference to figure 9 (e), and contain a lot pseudo-change informations among Fig. 9 (b), and have a lot because the assorted point that noise causes.The change-detection result that the present invention obtains in second experiment is shown in Fig. 9 (c), the change-detection result that the markov random file method obtains is shown in Fig. 9 (d), from Fig. 9 (c) as can be seen, the present invention approaches with reference to figure 9 (e) very much, and still has a lot of pseudo-change informations among Fig. 9 (d).From table 1 and table 2 as can be seen, the change-detection result that obtains of the present invention can both well reduce pseudo-change information in first experiment and second experiment two, and the minimizing false-alarm number has reduced the omission number, reduce total wrong number widely, thereby improved correct verification and measurement ratio.In first experiment, the change-detection result's that the present invention obtains in first group of data false-alarm number has reduced by 1650 pixels than the change-detection result's of differential technique method false-alarm number, total wrong number has reduced by 1635 pixels, the change-detection result's that the present invention obtains in second group of data false-alarm number has reduced by 5515 pixels than the change-detection result's of differential technique method false-alarm number, total wrong number has reduced by 5122 pixels, the change-detection result's that the present invention obtains in the 3rd group of data false-alarm number has reduced by 330 pixels than the change-detection result's of differential technique method false-alarm number, total wrong number has reduced by 543 pixels, illustrate that the present invention detects area-of-interest, and only classification can reduce false drop rate effectively to area-of-interest, improves the change-detection precision.In second experiment, the change-detection result's that the present invention obtains in first group of data false-alarm number has reduced by 1283 pixels than the change-detection result's of markov random file method false-alarm number, total wrong number has reduced by 1199 pixels, the change-detection result's that the present invention obtains in second group of data false-alarm number has reduced by 3569 pixels than the change-detection result's of markov random file method false-alarm number, total wrong number has reduced by 2999 pixels, the change-detection result's that the present invention obtains in the 3rd group of data false-alarm number has reduced by 283 pixels than the change-detection result's of markov random file method false-alarm number, total wrong number has reduced by 496 pixels, has illustrated that our bright consideration spatial neighborhood relation can access change-detection result more accurately.As can be seen, the present invention adopts that total wrong number, false-alarm number have obtained significantly reducing among the change-detection result that change-detection result that maximum variance between clusters obtains obtains than the markov random file method, and validity of the present invention has been described from table 1 and table 2.
Table 1: the change-detection evaluation of result index of three groups of experimental datas in first experiment
Table 2: the change-detection evaluation of result index of three groups of experimental datas in second experiment
Claims (5)
1. the method for detecting change of remote sensing image based on area-of-interest comprises the steps:
1) to the input two width of cloth not simultaneously the phase remote sensing images carry out the medium filtering denoising respectively;
The image corresponding pixel points gray scale difference value of phase calculates during 2) to filtered two, obtains a width of cloth differential chart;
3) the Canny edge of calculated difference figure, and calculate the characteristics of mean of each marginal point, obtain a breadths edge average figure;
4) with the fuzzy C Mean Method edge average figure is divided into non-certainly variation class, unmarked class and changes class certainly, and calculate pixel value maximum in the unmarked class and change the mean value of the minimum pixel value in the class as high threshold certainly, calculate the low threshold value of mean value conduct of affirming pixel value maximum in the non-variation class and the minimum pixel value in the unmarked class, according to high threshold and low threshold value edge average figure is cut apart respectively, obtained the low threshold value outline map of a panel height threshold value outline map and a width of cloth;
5) according to dual threshold boundary chain connection, the disconnected edge in the link high threshold outline map, obtaining the edge is the boundary chain map interlinking of enclosed region;
6) enclosed region in the boundary chain map interlinking of gained is inwardly outwards respectively expanded two pixels respectively along the normal orientation at its edge, enclosed region after expanding is carried out the zone fills, and the enclosed region after will filling is as area-of-interest, with other zones as non-area-of-interest;
7) with the minimal error rate threshold method pixel in the differential chart is divided into and changes class and non-variation class, differential chart is upgraded, obtain new differential chart according to classification, locus and the neighborhood relationships of the pixel of area-of-interest in the differential chart and non-area-of-interest;
8) calculate the maximum between-cluster variance threshold value of new differential chart and its area-of-interest is divided into changes class and non-variation class, the figure as a result of change-detection to the end.
2. method for detecting change of remote sensing image according to claim 1, the wherein characteristics of mean of described each marginal point of calculating of step 3), calculating as follows:
Be the center with each marginal point in the Canny outline map of calculated difference figure (2a), the edge line segment length that marginal point is formed in interior 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° eight directions of 3 * 7 sliding windows; the line segment of getting edge line segment length maximum is as the edge line in this sliding window; the pairing direction of this edge line is as the edge direction of current marginal point; and, obtain new Canny outline map with the removal of the marginal point on the non-edge line in the moving window;
(2b) calculate the pixel average of new Canny outline map marginal point respectively, and average that will be bigger obtains a breadths edge average figure as the characteristics of mean of this marginal point on 3 * 7 moving window inward flange line both sides of differential chart correspondence position.
3. method for detecting change of remote sensing image according to claim 1, wherein step 5) is described according to dual threshold edge connection method, connects the disconnected edge in the high threshold outline map, and obtaining the edge is the edge connection layout of enclosed region, carries out as follows:
(3a) from low threshold value outline map, deduct the high threshold outline map and obtain the edge differential chart, and with the high threshold outline map as the edge connection layout;
(3b) the disconnected marginal point in the search edge connection layout in the edge differential chart, if in the edge differential chart, there is marginal point to be connected in eight neighborhoods of marginal point, in the edge connection layout, be marginal point then with the position mark of this neighborhood with disconnected edge line in the high threshold outline map;
(3c) according to recurrence track algorithm continuous repeating step (3b) in the edge differential chart, bring in constant renewal in the edge connection layout, edges all in obtaining the edge connection layout are enclosed region.
4. method for detecting change of remote sensing image according to claim 1, wherein classification, locus and the neighborhood relationships of the described pixel according to area-of-interest and non-area-of-interest of step (7) are upgraded differential chart respectively, carry out as follows:
(4a) whether the locus of judging pixel in the differential chart in area-of-interest, is that intermediate value in 5 * 5 sliding windows at center is as the difference of current pixel point if current pixel point not in area-of-interest, is then calculated with the current pixel point; Otherwise, current pixel point in area-of-interest, execution in step (4b);
(4b) whether the classification of judging pixel in the differential chart belongs to the variation class, if belong to the variation class, then will be that maximal value in its 5 * 5 sliding window of center is as the difference of current pixel point with the current pixel point; Otherwise this pixel belongs to non-variation class, keeps the initial value in the differential chart.
5. method for detecting change of remote sensing image according to claim 1, the maximum between-cluster variance threshold value of the new differential chart of the described calculating of step (8) and area-of-interest is divided into changes class and non-variation class wherein, be the threshold value T that tries to achieve new differential chart according to the maximum between-cluster variance threshold method, and with in the area-of-interest in the new differential chart greater than the pixel of threshold value T as changing class, otherwise as non-variation class.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100242940A CN101694718B (en) | 2009-10-13 | 2009-10-13 | Method for detecting remote sensing image change based on interest areas |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100242940A CN101694718B (en) | 2009-10-13 | 2009-10-13 | Method for detecting remote sensing image change based on interest areas |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101694718A true CN101694718A (en) | 2010-04-14 |
| CN101694718B CN101694718B (en) | 2012-01-04 |
Family
ID=42093687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100242940A Expired - Fee Related CN101694718B (en) | 2009-10-13 | 2009-10-13 | Method for detecting remote sensing image change based on interest areas |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101694718B (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101894372A (en) * | 2010-08-03 | 2010-11-24 | 新疆大学 | New noise-containing remote sensing image segmentation method |
| CN101950364A (en) * | 2010-08-30 | 2011-01-19 | 西安电子科技大学 | Remote sensing image change detection method based on neighbourhood similarity and threshold segmentation |
| CN101968885A (en) * | 2010-09-25 | 2011-02-09 | 西北工业大学 | Method for detecting remote sensing image change based on edge and grayscale |
| CN102096921A (en) * | 2011-01-10 | 2011-06-15 | 西安电子科技大学 | SAR (Synthetic Aperture Radar) image change detection method based on neighborhood logarithm specific value and anisotropic diffusion |
| CN102270299A (en) * | 2011-08-24 | 2011-12-07 | 复旦大学 | Edge connection algorithm realized in parallel based on breakpoints |
| CN102819838A (en) * | 2012-07-17 | 2012-12-12 | 北京市遥感信息研究所 | Hyperspectral remote sensing image change detection method based on multisource target characteristic support |
| CN103268613A (en) * | 2013-05-28 | 2013-08-28 | 北京林业大学 | Method for detecting sub-compartment forest resource by remote sensing and geographical information system technology |
| CN103279954A (en) * | 2013-05-21 | 2013-09-04 | 武汉中测晟图遥感技术有限公司 | Remote-sensing image change detecting method based on land utilization database |
| CN105095846A (en) * | 2014-09-28 | 2015-11-25 | 航天恒星科技有限公司 | Method and system for extracting region growing seed points based on remote sensing images and sea-land segmentation |
| CN106875385A (en) * | 2017-02-09 | 2017-06-20 | 广州中国科学院软件应用技术研究所 | A kind of high robust region intrusion detection algorithm |
| CN107203957A (en) * | 2017-05-19 | 2017-09-26 | 上海海洋大学 | A kind of safely outsourced searching method of multidate high-spectrum remote sensing |
| CN107358625A (en) * | 2017-06-12 | 2017-11-17 | 西安电子科技大学 | The High Resolution SAR image change detection method detected based on SPP Net and area-of-interest |
| CN107679518A (en) * | 2017-10-27 | 2018-02-09 | 深圳极视角科技有限公司 | A kind of detecting system |
| CN110705427A (en) * | 2019-09-25 | 2020-01-17 | 中国人民解放军61646部队 | Extraction processing method and device for remote sensing image target area |
| CN110853026A (en) * | 2019-11-16 | 2020-02-28 | 四创科技有限公司 | Remote sensing image change detection method integrating deep learning and region segmentation |
| CN111476723A (en) * | 2020-03-17 | 2020-07-31 | 哈尔滨师范大学 | Method for recovering lost pixels of remote sensing image with failed L andsat-7 scanning line corrector |
| CN111814616A (en) * | 2020-06-28 | 2020-10-23 | 枫烨(深圳)科技有限公司 | Automatic examination paper marking processing system without answer sheet and implementation method thereof |
| CN113362286A (en) * | 2021-05-24 | 2021-09-07 | 江苏星月测绘科技股份有限公司 | Natural resource element change detection method based on deep learning |
-
2009
- 2009-10-13 CN CN2009100242940A patent/CN101694718B/en not_active Expired - Fee Related
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101894372A (en) * | 2010-08-03 | 2010-11-24 | 新疆大学 | New noise-containing remote sensing image segmentation method |
| CN101950364A (en) * | 2010-08-30 | 2011-01-19 | 西安电子科技大学 | Remote sensing image change detection method based on neighbourhood similarity and threshold segmentation |
| CN101968885A (en) * | 2010-09-25 | 2011-02-09 | 西北工业大学 | Method for detecting remote sensing image change based on edge and grayscale |
| CN102096921B (en) * | 2011-01-10 | 2013-02-27 | 西安电子科技大学 | SAR (Synthetic Aperture Radar) image change detection method based on neighborhood logarithm specific value and anisotropic diffusion |
| CN102096921A (en) * | 2011-01-10 | 2011-06-15 | 西安电子科技大学 | SAR (Synthetic Aperture Radar) image change detection method based on neighborhood logarithm specific value and anisotropic diffusion |
| CN102270299A (en) * | 2011-08-24 | 2011-12-07 | 复旦大学 | Edge connection algorithm realized in parallel based on breakpoints |
| CN102819838A (en) * | 2012-07-17 | 2012-12-12 | 北京市遥感信息研究所 | Hyperspectral remote sensing image change detection method based on multisource target characteristic support |
| CN102819838B (en) * | 2012-07-17 | 2015-03-25 | 北京市遥感信息研究所 | Hyperspectral remote sensing image change detection method based on multisource target characteristic support |
| CN103279954A (en) * | 2013-05-21 | 2013-09-04 | 武汉中测晟图遥感技术有限公司 | Remote-sensing image change detecting method based on land utilization database |
| CN103268613A (en) * | 2013-05-28 | 2013-08-28 | 北京林业大学 | Method for detecting sub-compartment forest resource by remote sensing and geographical information system technology |
| CN103268613B (en) * | 2013-05-28 | 2016-06-08 | 北京林业大学 | Utilize the method that remote sensing, geographic information system technology detect bottom class's forest resourceies |
| CN105095846B (en) * | 2014-09-28 | 2019-07-30 | 航天恒星科技有限公司 | Region growing seed point extracting method and system towards the segmentation of remote sensing images sea land |
| CN105095846A (en) * | 2014-09-28 | 2015-11-25 | 航天恒星科技有限公司 | Method and system for extracting region growing seed points based on remote sensing images and sea-land segmentation |
| CN106875385A (en) * | 2017-02-09 | 2017-06-20 | 广州中国科学院软件应用技术研究所 | A kind of high robust region intrusion detection algorithm |
| CN107203957A (en) * | 2017-05-19 | 2017-09-26 | 上海海洋大学 | A kind of safely outsourced searching method of multidate high-spectrum remote sensing |
| CN107358625A (en) * | 2017-06-12 | 2017-11-17 | 西安电子科技大学 | The High Resolution SAR image change detection method detected based on SPP Net and area-of-interest |
| CN107679518A (en) * | 2017-10-27 | 2018-02-09 | 深圳极视角科技有限公司 | A kind of detecting system |
| CN110705427A (en) * | 2019-09-25 | 2020-01-17 | 中国人民解放军61646部队 | Extraction processing method and device for remote sensing image target area |
| CN110853026A (en) * | 2019-11-16 | 2020-02-28 | 四创科技有限公司 | Remote sensing image change detection method integrating deep learning and region segmentation |
| CN110853026B (en) * | 2019-11-16 | 2023-04-07 | 四创科技有限公司 | Remote sensing image change detection method integrating deep learning and region segmentation |
| CN111476723A (en) * | 2020-03-17 | 2020-07-31 | 哈尔滨师范大学 | Method for recovering lost pixels of remote sensing image with failed L andsat-7 scanning line corrector |
| CN111476723B (en) * | 2020-03-17 | 2023-04-18 | 哈尔滨师范大学 | Remote sensing image lost pixel recovery method for failure of Landsat-7 scanning line corrector |
| CN111814616A (en) * | 2020-06-28 | 2020-10-23 | 枫烨(深圳)科技有限公司 | Automatic examination paper marking processing system without answer sheet and implementation method thereof |
| CN113362286A (en) * | 2021-05-24 | 2021-09-07 | 江苏星月测绘科技股份有限公司 | Natural resource element change detection method based on deep learning |
| CN113362286B (en) * | 2021-05-24 | 2022-02-01 | 江苏星月测绘科技股份有限公司 | Natural resource element change detection method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101694718B (en) | 2012-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101694718B (en) | Method for detecting remote sensing image change based on interest areas | |
| Ouma et al. | Pothole detection on asphalt pavements from 2D-colour pothole images using fuzzy c-means clustering and morphological reconstruction | |
| Grinias et al. | MRF-based segmentation and unsupervised classification for building and road detection in peri-urban areas of high-resolution satellite images | |
| CN106203398B (en) | A kind of method, apparatus and equipment detecting lane boundary | |
| Valero et al. | Advanced directional mathematical morphology for the detection of the road network in very high resolution remote sensing images | |
| CN100545867C (en) | Aerial shooting traffic video frequency vehicle rapid checking method | |
| CN102509098B (en) | Fisheye image vehicle identification method | |
| CN108052904B (en) | Method and device for acquiring lane line | |
| CN103699905B (en) | Method and device for positioning license plate | |
| CN106651872A (en) | Prewitt operator-based pavement crack recognition method and system | |
| CN103366156A (en) | Road structure detection and tracking | |
| CN104809433A (en) | Zebra stripe detection method based on maximum stable region and random sampling | |
| CN103310435B (en) | Combine vertical projection and optimal path the method split characters on license plate | |
| CN110334634A (en) | A kind of detection method and prior-warning device of lane line classification | |
| CN101950364A (en) | Remote sensing image change detection method based on neighbourhood similarity and threshold segmentation | |
| CN103413303A (en) | Infrared target segmentation method based on joint obviousness | |
| CN102819952A (en) | Method for detecting illegal lane change of vehicle based on video detection technique | |
| CN104268872A (en) | Consistency-based edge detection method | |
| CN109635733B (en) | Parking lot and vehicle target detection method based on visual saliency and queue correction | |
| CN103383733A (en) | Lane video detection method based on half-machine study | |
| CN101634705B (en) | Method for detecting target changes of SAR images based on direction information measure | |
| CN108171695A (en) | A kind of express highway pavement detection method based on image procossing | |
| CN109886168B (en) | Ground traffic sign identification method based on hierarchy | |
| CN112001216A (en) | Automobile driving lane detection system based on computer | |
| CN105117726A (en) | License plate positioning method based on multi-feature area accumulation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120104 Termination date: 20171013 |