CN102930269A - Method for modulating multiscale semanteme of remote-sensing image - Google Patents
Method for modulating multiscale semanteme of remote-sensing image Download PDFInfo
- Publication number
- CN102930269A CN102930269A CN2012103444537A CN201210344453A CN102930269A CN 102930269 A CN102930269 A CN 102930269A CN 2012103444537 A CN2012103444537 A CN 2012103444537A CN 201210344453 A CN201210344453 A CN 201210344453A CN 102930269 A CN102930269 A CN 102930269A
- Authority
- CN
- China
- Prior art keywords
- sensing image
- remote sensing
- pixel
- scale
- yardstick
- 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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a method for modulating multiscale semanteme of a remote-sensing image. The method comprises the following steps of: selecting local feature types of processing objects of the remote-sensing image; determining the size range of observation scales which are used by the processing objects of the remote-sensing image; partitioning the levels of the observation scales of the processing objects of the remote-sensing image; generating single-pixel feature value sequences; generating multiscale feature values of pixels; and generating a scale manifesting image. According to the method, the internal consistency of the processing objects of the remote-sensing image is improved and the relative property between the processing objects of the remote-sensing image and a background is enhanced by utilizing the acquired scale manifesting image modulated by the principle. Test results prove that the multiscale feature-based scale manifesting image can reflect the actual condition of surface features well, so the method has high using value.
Description
Technical field
The present invention relates to the modulator approach of the multiple dimensioned semanteme of a kind of remote sensing image, belong to the remote sensing image preconditioning technique towards remote sensing image ground object target identification, can be applicable to the extraction of the ground object targets such as road, buildings based on remote sensing image.
Background technology
In the remote sensing image information extractive technique field, yardstick has different implications, at first refers to the geometric scale of the described true atural object of remote sensing image, and this geometric scale was ignorant before can obtaining terrestrial object information by remote sensing image; Next is to observe yardstick, this observation yardstick shows as the resolution of remote sensing image at original remote sensing image, the remote sensing image of observing yardstick little (high resolving power) can generate the remote sensing image of observing yardstick large (low resolution) according to image-forming principle, thereby high-resolution remote sensing image is that multiple dimensioned observational study provides the foundation; The yardstick in Remote Sensing Image Segmentation research at last, the remote sensing image object size that yardstick refers generally to cut apart (or parameter of control cutting unit size), the parameter of control cutting unit size is generally inner consistent degree parameter, although exist close relationship between the consistent cell size of spending and cutting apart, but be difficult to clearly describe, this also is the aspect that is difficult to control most in the eCognition software, must obtain suitable scale parameter value by repetition test.
All characteristics of remote sensing images all are the features under certain yardstick, it is the yardstick characteristics of remote sensing image, generally referred to as characteristics of remote sensing image or feature, when a plurality of yardstick remote sensing image of research, be referred to as Analysis On Multi-scale Features, but carry out respectively often under study for action the feature generation net result that single scale research considers each yardstick at last again.Fractal object has feature not with the characteristic of dimensional variation; More general situation is that certain feature only can remain unchanged in the some scale scope, and the SIFT method has been used this feature; The atural object relation is complicated on the remote sensing image, and feature changes along with the variation of yardstick, and closely related with its tax dis environment.
In the multiple dimensioned remote sensing image identification patent documentation that relates to, its disposal route generally is to identify under each single scale image, at last the recognition result under the different scale condition is merged; In the multiple dimensioned Remote Sensing Image Segmentation that relates to, the order that generally adopts yardstick to increase progressively produces the segmentation result of different scale successively, at last the segmentation result under the different scale condition is merged, can be referring to 200710304466.0,200810032390.5,201110137557.6, as shown in Figure 1.In above-mentioned patent, when using yardstick to analyze, scale feature lies in the remote sensing image context, and what use in application is single scale feature, and does not consider multiple dimensioned variation characteristic.And as high-resolution remote sensing image, the precision height will reflect the detail characteristic of remote sensing image on the one hand, the zone of remote sensing image is all larger on the other hand, thereby has multiple dimensioned characteristic on the remote sensing image, multiple dimensioned characteristic is the important evidence of remote sensing target identification, if the yardstick characteristic image clear that generates can not have been described this scale feature by explicitly, then can affect the figure image intensifying ability of remote sensing image.
Summary of the invention
The technology of the present invention is dealt with problems: overcome the deficiencies in the prior art, the modulator approach of the multiple dimensioned semanteme of a kind of remote sensing image is provided, by a kind of multiple dimensioned semantic modulator approach that the remote sensing image scale feature is clear in the pixel of remote sensing image, generate yardstick image clear, the yardstick image explicitly clear that generates has been described the remote sensing image scale feature, can reflect the dimensional variation feature of remote sensing image, belong to the high-order characteristics of remote sensing image, be used for the identification of high-level semantic remote sensing image object.
The technical solution adopted for the present invention to solve the technical problems is as follows: the modulator approach of the multiple dimensioned semanteme of a kind of remote sensing image, and implementation step is as follows:
(1) selected remote sensing image is processed the local feature type of object and selected remote sensing image processing object; Described remote sensing image is processed object and is referred in remote sensing image an interested pending zone; The local feature that remote sensing image is processed object is reflected in the relative characteristic that this remote sensing image is processed adjacent remote sensing image processing object with other in the subrange of object, the local feature that remote sensing image is processed object also is reflected in this remote sensing image and processes between the set of pixels that object comprises and adjacent remote sensing image with other is processed relative characteristic between the pixel that object comprises, utilizes local feature information that remote sensing image processes object as judging that remote sensing image processes the similarity foundation between object;
(2) on the basis of selected remote sensing image processing object, process the big or small size of determining the observation yardstick of employed remote sensing image processing object of geometric scale of object according to selected remote sensing image;
(3) adopt the scale feature curve to express in a continuous manner remote sensing image and process the local feature situation of change of pixel in observing range scale that object comprises; Adopt discrete form to sample and encode for the scale feature curve, the observation scale size scope of the definite remote sensing image of institute in the step 2 being processed object is divided into n observation yardstick according to m times of progression;
(4) process object local feature type according to the remote sensing image of step 1, calculate n the local feature of observing under the yardstick that each pixel is divided in step 3 in the remote sensing image, each pixel is observed single pixel characteristic value sequence that the local feature that obtains under the yardstick becomes this pixel at n in the remote sensing image; Each pixel has single pixel characteristic value sequence of this pixel in the remote sensing image;
(5) need to encode to single pixel characteristic value sequence of each pixel in the remote sensing image that obtains in the step 4 and be summarised as an eigenwert of this pixel, Analysis On Multi-scale Features value as this pixel in the remote sensing image, to describe the Analysis On Multi-scale Features on the remote sensing image pixel, a pixel Analysis On Multi-scale Features value can reflect that pixel at the local feature of certain observation yardstick, also can reflect the local feature of this pixel on each yardstick of observing range scale simultaneously in the remote sensing image;
(6) in the step 5, each pixel generates an Analysis On Multi-scale Features value in the remote sensing image, with the Analysis On Multi-scale Features value of this pixel pixel value as this pixel, it is identical with the original remote sensing image size to form a width of cloth, the remote sensing image of the Analysis On Multi-scale Features value that this pixel value shows in original remote sensing image for this pixel, i.e. yardstick image clear.
The scope that employing remote sensing image in the described step (1) is processed the observation scale size of object is 2 times of the remote sensing image scope of processing object geometric scale size.
M=2 in the described step (3).
The present invention's advantage compared with prior art is:
(1) during traditional multiple dimensioned remote sensing image is processed, is that a plurality of yardsticks are carried out processing towards the remote sensing image of final goal, and with each result simple combination, can't uses the dimensional variation feature; The yardstick characteristic image explicitly clear that the present invention generates has been described this feature, can reflect the dimensional variation feature of remote sensing image, belongs to the high-order characteristics of remote sensing image, can be used for the identification of high-level semantic object.
(2) yardstick of the present invention image clear with yardstick as the modulation framework, process the characteristics of objects type at remote sensing image and elect as and carry out the method for the invention under the bright dark characteristic type and modulate the yardstick that obtains image clear, can keep the spectral characteristic of raw video.
(3) during traditional multiple dimensioned remote sensing image was processed, the feature under the large scale was because decrease resolution often departs from real location point; Described in step 6, and the feature on the yardstick of the present invention image clear all is based on each pixel generation Analysis On Multi-scale Features value in the original remote sensing image, forms yardstick image clear, thereby has kept the positional precision of original remote sensing image.
(4) described in step 4 and step 5, each pixel value is this pixel Analysis On Multi-scale Features value in the yardstick of the present invention characteristic image clear, to be summarized by the single scale characteristic sequence of this pixel to form, can when summarizing the single scale characteristic sequence, select very easily part or specific several yardstick to be summarised as the Analysis On Multi-scale Features value of this pixel, thereby realize reinforcement or inhibition to yardstick, with the spatial contrast degree of outstanding certain yardstick object, this is difficult to accomplish in classic method.
(5) reflection of the pixel value in the yardstick of the present invention image clear is the multiple dimensioned performance of this pixel in raw video, thereby similarity between neighbor has reflected whether these pixels belong to same remote sensing image processing object in original remote sensing image in the yardstick image clear, Analysis On Multi-scale Features in the yardstick of the present invention image clear can be applied to accordingly classify or cut apart in the research, this is a kind of new similarity definition, can remedy that to lack the dimensional variation similarity in the classic method not enough, make cut apart or nicety of grading higher.
Description of drawings
Fig. 1 is common scale processing process synoptic diagram;
Fig. 2 is for using scale processing process synoptic diagram of the present invention;
Fig. 3 is yardstick of the present invention image modulated process clear process flow diagram;
Fig. 4 is the spectrum section of object of the present invention in the neighborhood scope, and wherein W is that pixel or remote sensing image are processed object, and A, B, C are its different Size of Neighborhoods;
Fig. 5 is the relative brightness-scale feature change curve of object of the present invention;
Fig. 6 is modulation lamppost model among the present invention;
Fig. 7 is yardstick of the present invention image effect figure clear, and the picture left above is original remote sensing image, and top right plot is yardstick corresponding to the picture left above image clear, and lower-left figure is another width of cloth original remote sensing image, and bottom-right graph is yardstick corresponding to lower-left figure image clear.Can find out that therefrom yardstick image energy clear is better expressed remote sensing image and processed object.
Embodiment
The invention belongs to the remote sensing image preconditioning technique that extracts towards the remote sensing image ground object target, be used for remote sensing image clear and process the space scale meaning of one's words of object, it is clear in the pixel of remote sensing image with the remote sensing image scale feature by a kind of multiple dimensioned semantic modulator approach, generate yardstick image clear, the yardstick image explicitly clear that generates has been described the remote sensing image scale feature, the dimensional variation feature that can reflect remote sensing image, belong to the high-order characteristics of remote sensing image, can be used for the identification of high-level semantic remote sensing image object.
Yardstick image modulated process clear process flow diagram such as Fig. 3.Can express the local feature type that remote sensing image is processed object comprising selecting, processing object for concrete remote sensing image determines to observe the scale size scope and observes the yardstick grade, observe extraction local feature value under the yardstick in difference, form the single scale characteristic sequence of pixel, and the single scale characteristic value sequence summarized the scale feature value that forms pixel, form at last yardstick image clear.
Shown in Fig. 2,3, concrete steps of the present invention are as follows.
1. select remote sensing image to process the local feature type of object
Selected remote sensing image is processed the local feature type of object and selected remote sensing image processing object; The local feature that remote sensing image is processed object is reflected in the relative characteristic that this remote sensing image is processed adjacent remote sensing image processing object with other in the subrange of object, the local feature that remote sensing image is processed object also is reflected in this remote sensing image and processes between the set of pixels that object comprises and adjacent remote sensing image with other is processed relative characteristic between the pixel that object comprises, utilizes local feature information that remote sensing image processes object as judging that remote sensing image processes the similarity foundation between object.Local feature is divided into consistance characteristics and comparison feature, the consistance feature is that remote sensing image is processed the common trait that has between object in the certain area in remote sensing image, such as average, deviation, texture etc., the contrast characteristic is that other remote sensing image processes that object compares and the feature that presents in remote sensing image is processed object and neighborhood on every side, such as light and shade, number, foreground-background, linear direction etc.
2. determine the observation scale size scope that remote sensing image processing object adopts
On the basis of the selected remote sensing image processing object of observing, determine the size of the observation yardstick of employed remote sensing image processing object according to the geometric scale size of selected remote sensing image processing object; The scope that remote sensing image that the present invention adopts is processed the observation scale size of object is 2 times of the remote sensing image scope of processing object geometric scale size; This step is processed the observation scale size scope that object geometric scale size is determined remote sensing image processing object mainly for remote sensing image.Observe a remote sensing image and process object, the observation scale size of remote sensing image processing object is greater than the geometric scale that this remote sensing image is processed object itself at least so.It is too little to observe range scale, can not reflect the geometric scale size of remote sensing image processing object; It is too large to observe the scale size scope, then be not enough to this remote sensing image of accurate expression and process the interior local feature of subrange of object, thereby the present invention considers and can process under the object local feature prerequisite by the accurate expression remote sensing image, consider simultaneously the problem of implementation efficiency, the observation scale size scope of determining remote sensing image processing object is about 2 times of its geometric scale magnitude range, the minimum value of namely observing yardstick is that remote sensing image is processed half of object detail characteristic yardstick, maximal value is 2 times that whole remote sensing image is processed the object geometric scale, process object for unlimited extension remote sensing image, road for example, with its width as the geometric scale foundation.
3. divide remote sensing image and process object observation yardstick rank
Adopt the scale feature curve to express in a continuous manner remote sensing image and process the local feature situation of change of pixel in observing range scale that object comprises; A pixel processing the pixel that object was comprised with remote sensing image illustrates that as shown in Figure 4 wherein P, Q, W are that pixel or remote sensing image are processed object, and A, B, C are its different Size of Neighborhoods.A bit for its neighborhood, can show as bright or dark, and scope is different, and it is bright secretly also different, the position of W shows as in the A neighborhood secretly among the figure, in the B neighborhood, then show as bright, then show as dark in the wider C neighborhood, the space scale change to attributes of this point has just intactly been expressed in the combination of this " dark-light-dark ", compares with simple pixel value and has expressed more semantic relation, can be understood as this point and be a relatively bright point in the blackening in the clear zone.Fig. 5 display pixel point is its relative change curve of brightness in its neighborhood under different scale.In discrete scale processing, will obtain a scale feature sequence.
The scale feature curve can fine expression pixel difference in the observing range scale continuous situation of change of observing the yardstick lower eigenvalue, but in computing machine, only can adopt discrete form, thereby need to sample and encode, namely in observing range scale, divide and observe the yardstick rank.The present invention processes object at definite remote sensing image and observes on the scale size scope basis, and the observation scale size scope of the definite remote sensing image of institute in the step 2 being processed object is divided into n observation yardstick according to 2 times of progression.Be convenient to like this expression and the post-processed of remote sensing image.Be about to observe metric space and be divided into n rank, higher leveled observation scale size is 2 times of observation scale size of adjacent low one-level, be divided in like this and be convenient in the operation realize, and in theory, after the observation yardstick was 2 times of sizes of previous observation yardstick, the remote sensing image that reflects that more can be accurate and don't redundant was processed object geometric scale information.
4. generate single pixel characteristic value sequence
Process object local feature type according to the remote sensing image of step 1, calculate n the local feature of observing under the yardstick that each pixel is divided in step 3 in the remote sensing image, each pixel is observed single pixel characteristic value sequence that the local feature that obtains under the yardstick becomes this pixel at n in the remote sensing image; Each pixel has single pixel characteristic value sequence of this pixel in the remote sensing image.
5. generate pixel Analysis On Multi-scale Features value
Need to encode to single pixel characteristic value sequence of each pixel in the remote sensing image that obtains in the step 4 and to be summarised as an eigenwert of this pixel, Analysis On Multi-scale Features value as this pixel in the remote sensing image, to describe the Analysis On Multi-scale Features on the remote sensing image pixel, a pixel Analysis On Multi-scale Features value can reflect that pixel at the local feature of certain observation yardstick, also can reflect the local feature of this pixel on each yardstick of observing range scale simultaneously in the remote sensing image.The coded representation that the present invention expresses the Analysis On Multi-scale Features value is:
Wherein, S is the Analysis On Multi-scale Features value of a pixel of formation, i.e. global feature value; K is for observing yardstick rank k=1 ... n; t
kBe k the switch of observing yardstick; w
kBe k and observe the yardstick weight, observe the yardstick weight and reflected that the scale feature that shows under the different observations is on the impact of global feature value; f
kBe that k observes the scale feature value under the yardstick, the observation feature of reflection pixel under k observation yardstick.When the weight of adjacent scale feature is 2 times when concerning, w
k=2
k, can feature under the single observation yardstick be described with a bit position, realized the explicit description of feature under the single observation yardstick, the global feature value is then described multiple dimensioned comprehensive characteristics.
6. generate yardstick image clear
In step 5, each pixel generates an Analysis On Multi-scale Features value in the remote sensing image, with the Analysis On Multi-scale Features value of this pixel pixel value as this pixel, it is identical with the original remote sensing image size to form a width of cloth, but the remote sensing image of the Analysis On Multi-scale Features value that pixel value shows in original remote sensing image for this pixel, i.e. yardstick image clear.Why be referred to as image clear and be because the pixel value of each pixel has not been simple radiance value or relative brightness value, but a kind ofly react this and put eigenwert under the multiple dimensioned observation.
To the as above process of 6 steps, be referred to as modulated process, be about to the pixel value of primitive reaction terrestrial radiation intensity, from the gray scale codomain, the observations (single scale feature) of observing under the yardstick by difference is modulated to the Analysis On Multi-scale Features codomain.In realization, observe and adopt the equal proportion relation between the yardstick, the weight ratio of adjacent observation scale feature value is selected 2 times of relations, and the modulated process of such yardstick characteristic image clear can pass through the lamppost model realization.Modulator approach based on bright dark feature is: as shown in Figure 6, imagine a pillar that stands in pixel center, the binary number of original pixel value is arranged in order from top to bottom, in each binary digit a lamp is set, the scope that illuminates of every lamp is adjacent lamps must illuminate scope below it 2 times, if corresponding binary digit is 1, then this grade is opened, otherwise this lamp is closed, and then total brightness represents the primary radiation gray-scale value.During modulation, to the opening and closing based on following judgment principle of lamp on each binary digit, if namely the gray-scale value of this point belongs to bright pixel in the ranges of exposures such as this position, then turn on this lamp, otherwise close.Binary bit sequence reflection after the modulation is the Analysis On Multi-scale Features based on bright dark feature on this aspect like this.The modulator approach of further feature is identical.
In the lamppost model, scale-of-two pixel value representation on the yardstick image clear the image feature of each yardstick level, only turn on one deck lamp (value that only keeps this by masking operation) then yardstick image display clear be the characteristics of remote sensing image of under this observation yardstick, observing.When high-order lamp is extinguished successively, the observation yardstick of the characteristics of remote sensing image that yardstick image clear is described will reduce gradually, only see at last micro-texture, and lose whole object; Otherwise little texture disappears, and large object highlights.
Yardstick image clear has strengthened difference between the ground object target object by considering Analysis On Multi-scale Features, can be used for the more accurate Remote Sensing Image Segmentation that carries out, similarity measurement in the cutting procedure can be comprehensive Analysis On Multi-scale Features value, also can be the dimensional variation form that the Analysis On Multi-scale Features position is described.During as the measuring similarity standard, the result of cutting apart has the light and shade consistance of comprehensive observing with comprehensive Analysis On Multi-scale Features value; During as the measuring similarity standard, can be classified as a class by the pixel that the scale feature metamorphosis is consistent with multiple dimensioned change shape, realize that the similar remote sensing image of different scale processes the unification of object.
The yardstick image effect clear that utilizes principle of the invention modulation to obtain is shown in Figure 7, the picture left above is original remote sensing image, top right plot is yardstick corresponding to the picture left above image clear, and lower-left figure is another width of cloth original remote sensing image, and bottom-right graph is yardstick corresponding to lower-left figure image clear.Therefrom can find out, yardstick image energy clear is better expressed remote sensing image and is processed object, can find out that therefrom yardstick image energy clear better improves remote sensing image to be processed the internal consistency of object, strengthen the target that remote sensing image is processed the contrast of object and Background.Experiment results proved based on the yardstick of Analysis On Multi-scale Features image clear more can reasonable reflection atural object actual conditions, show that the present invention has very strong use value.
The yardstick image effect clear that utilizes principle of the invention modulation to obtain can better improve remote sensing image to be processed the internal consistency of object, strengthens the target that remote sensing image is processed the contrast of object and Background.Experiment results proved based on the yardstick of Analysis On Multi-scale Features image clear more can reasonable reflection atural object actual conditions, show that the present invention has very strong use value.
The content that is not described in detail in the instructions of the present invention belongs to the known prior art of this area professional and technical personnel.
The above only is preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (3)
1. the modulator approach of the multiple dimensioned semanteme of remote sensing image is characterized in that performing step is as follows:
(1) selected remote sensing image is processed the local feature type of object and selected remote sensing image processing object; Described remote sensing image is processed object and is referred in remote sensing image an interested pending zone; The local feature that remote sensing image is processed object is reflected in the relative characteristic that this remote sensing image is processed adjacent remote sensing image processing object with other in the subrange of object, the local feature that remote sensing image is processed object also is reflected in this remote sensing image and processes between the set of pixels that object comprises and adjacent remote sensing image with other is processed relative characteristic between the pixel that object comprises, utilizes local feature information that remote sensing image processes object as judging that remote sensing image processes the similarity foundation between object;
(2) on the basis of selected remote sensing image processing object, process the big or small size of determining the observation yardstick of employed remote sensing image processing object of geometric scale of object according to selected remote sensing image;
(3) adopt the scale feature curve to express in a continuous manner remote sensing image and process the local feature situation of change of pixel in observing range scale that object comprises; Adopt discrete form to sample and encode for the scale feature curve, the observation scale size scope of the definite remote sensing image of institute in the step 2 being processed object is divided into n observation yardstick according to m times of progression;
(4) process object local feature type according to the remote sensing image of step 1, calculate n the local feature of observing under the yardstick that each pixel is divided in step 3 in the remote sensing image, each pixel is observed single pixel characteristic value sequence that the local feature that obtains under the yardstick becomes this pixel at n in the remote sensing image; Each pixel has single pixel characteristic value sequence of this pixel in the remote sensing image;
(5) need to encode to single pixel characteristic value sequence of each pixel in the remote sensing image that obtains in the step 4 and be summarised as an eigenwert of this pixel, Analysis On Multi-scale Features value as this pixel in the remote sensing image, to describe the Analysis On Multi-scale Features on the remote sensing image pixel, a pixel Analysis On Multi-scale Features value can reflect that pixel at the local feature of certain observation yardstick, also can reflect the local feature of this pixel on each yardstick of observing range scale simultaneously in the remote sensing image;
(6) in the step 5, each pixel generates an Analysis On Multi-scale Features value in the remote sensing image, with the Analysis On Multi-scale Features value of this pixel pixel value as this pixel, it is identical with the original remote sensing image size to form a width of cloth, the remote sensing image of the Analysis On Multi-scale Features value that this pixel value shows in original remote sensing image for this pixel, i.e. yardstick image clear.
2. the modulator approach of the multiple dimensioned semanteme of remote sensing image according to claim 1 is characterized in that: the scope that the employing remote sensing image in the described step (1) is processed the observation scale size of object is 2 times of the remote sensing image scope of processing object geometric scale size.
3. the modulator approach of the multiple dimensioned semanteme of remote sensing image according to claim 1 is characterized in that: the m=2 in the described step (3).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210344453.7A CN102930269B (en) | 2012-09-17 | 2012-09-17 | Method for modulating multiscale semanteme of remote-sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210344453.7A CN102930269B (en) | 2012-09-17 | 2012-09-17 | Method for modulating multiscale semanteme of remote-sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102930269A true CN102930269A (en) | 2013-02-13 |
| CN102930269B CN102930269B (en) | 2015-05-13 |
Family
ID=47645066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210344453.7A Expired - Fee Related CN102930269B (en) | 2012-09-17 | 2012-09-17 | Method for modulating multiscale semanteme of remote-sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102930269B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109376599A (en) * | 2018-09-19 | 2019-02-22 | 中国科学院东北地理与农业生态研究所 | A kind of remote sensing image processing method and system extracted towards wetland information |
| CN110096948A (en) * | 2019-03-15 | 2019-08-06 | 中国科学院西安光学精密机械研究所 | Remote sensing image recognition methods based on characteristic aggregation convolutional network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060083418A1 (en) * | 2003-02-11 | 2006-04-20 | Qinetiq Limited | Image analysis |
| CN102169545A (en) * | 2011-04-25 | 2011-08-31 | 中国科学院自动化研究所 | Detection method for changes of high-resolution remote sensing images |
| CN102496151A (en) * | 2011-12-08 | 2012-06-13 | 南京大学 | Method for multi-scale segmentation of high-resolution remote sensing images |
-
2012
- 2012-09-17 CN CN201210344453.7A patent/CN102930269B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060083418A1 (en) * | 2003-02-11 | 2006-04-20 | Qinetiq Limited | Image analysis |
| CN102169545A (en) * | 2011-04-25 | 2011-08-31 | 中国科学院自动化研究所 | Detection method for changes of high-resolution remote sensing images |
| CN102496151A (en) * | 2011-12-08 | 2012-06-13 | 南京大学 | Method for multi-scale segmentation of high-resolution remote sensing images |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109376599A (en) * | 2018-09-19 | 2019-02-22 | 中国科学院东北地理与农业生态研究所 | A kind of remote sensing image processing method and system extracted towards wetland information |
| CN110096948A (en) * | 2019-03-15 | 2019-08-06 | 中国科学院西安光学精密机械研究所 | Remote sensing image recognition methods based on characteristic aggregation convolutional network |
| CN110096948B (en) * | 2019-03-15 | 2020-11-17 | 中国科学院西安光学精密机械研究所 | Remote sensing image identification method based on characteristic aggregation convolutional network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102930269B (en) | 2015-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shendryk et al. | Deep learning for multi-modal classification of cloud, shadow and land cover scenes in PlanetScope and Sentinel-2 imagery | |
| Ling et al. | Land cover change mapping at the subpixel scale with different spatial-resolution remotely sensed imagery | |
| Zhou et al. | Automated residential building detection from airborne LiDAR data with deep neural networks | |
| Hall et al. | Detecting dominant landscape objects through multiple scales: An integration of object-specific methods and watershed segmentation | |
| Yang et al. | An automated method to parameterize segmentation scale by enhancing intrasegment homogeneity and intersegment heterogeneity | |
| CN107067405B (en) | Remote sensing image segmentation method based on scale optimization | |
| CN106022341B (en) | Water body information method and system after high-resolution optical remote sensing image calamity | |
| JP5103665B2 (en) | Object tracking device and object tracking method | |
| Kalkan et al. | Comparison of support vector machine and object based classification methods for coastline detection | |
| Pathak | New Change Detection Techniques to monitor land cover dynamics in mine environment | |
| Tang et al. | A multiple-point spatially weighted k-NN method for object-based classification | |
| US20210027055A1 (en) | Methods and Systems for Identifying Topographic Features | |
| CN114841972A (en) | Power transmission line defect identification method based on saliency map and semantic embedded feature pyramid | |
| Liang et al. | An extraction and classification algorithm for concrete cracks based on machine vision | |
| CN107229910A (en) | A kind of remote sensing images icing lake detection method and its system | |
| CN102930269B (en) | Method for modulating multiscale semanteme of remote-sensing image | |
| O’Neil-Dunne et al. | Incorporating contextual information into object-based image analysis workflows | |
| Lizarazo et al. | Fuzzy image segmentation for urban land-cover classification | |
| Kruse et al. | Marked point processes for the automatic detection of bomb craters in aerial wartime images | |
| Ma et al. | Urban landscape classification using Chinese advanced high-resolution satellite imagery and an object-oriented multi-variable model | |
| Ennouri et al. | Detection of urban and environmental changes via remote sensing | |
| Kruse et al. | Generating impact maps from automatically detected bomb craters in aerial wartime images using marked point processes | |
| Zhang et al. | Application and evaluation of object-oriented technology in high-resolution remote sensing image classification | |
| Zhang et al. | An optimal approach for crack extraction from UAV sub-images after cutting | |
| Kupidura | Semi-automatic method for a built-up area intensity survey using morphological granulometry |
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 |
Granted publication date: 20150513 Termination date: 20190917 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |