CN102915524B - Method for eliminating shadow based on match of inside and outside check lines of shadow area - Google Patents
Method for eliminating shadow based on match of inside and outside check lines of shadow area Download PDFInfo
- Publication number
- CN102915524B CN102915524B CN201210342195.9A CN201210342195A CN102915524B CN 102915524 B CN102915524 B CN 102915524B CN 201210342195 A CN201210342195 A CN 201210342195A CN 102915524 B CN102915524 B CN 102915524B
- Authority
- CN
- China
- Prior art keywords
- shadow
- grey scale
- line
- curve
- scale curve
- 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.)
- Active
Links
Landscapes
- Image Processing (AREA)
Abstract
A method for eliminating a shadow based on match of inside and outside check lines of a shadow area comprises the steps that: the shadow area is obtained through shadow detection so as to generate the inside and outside check lines of the shadow area; for each shadow area, the corresponding grayscale curves of the inside and outside check lines are obtained through waveband iteration, the radiation difference between the shadow area and the periphery can be obtained through matching the grayscale curves, and radiation correction is carried out on the shadow area according to the radiation difference until the shadow area inside and outside radiation difference which is reflected by the corresponding luminance curves of the inside and outside check lines is small enough; and a transition area is built according to the inside and outside check lines and is feathered. The method for eliminating shadow is simple, has strong operability, not only can eliminate the macroscopic radiation difference between the shadow area and peripheral land features, but also is used for solving the detail transition problem of shadow boundaries, so that a good visual effect is obtained after eliminating the image shadows.
Description
Technical field
The present invention relates to Remote Sensing Image Processing Technology field, especially relating to a kind of shadow removing method based on checking lines matching inside and outside shadow region.
Background technology
Shade is the ubiquitous a kind of physical phenomenon of occurring in nature, and in remote sensing images, the existence of shade can cause adverse influence to the relevant issues of the computer visions such as object identification and remote sensing image subsequent treatment.Shade phenomenon is also a kind of special image degradation form, the target information in shadow region can be caused to be interfered or to lose, be unfavorable for the atural object edge extracting in Computer Image Processing, target identification, target tracking and ground mulching classification and affect the success ratio of matching algorithm, impact is detected for the change based on remote sensing image larger.Consistent in order to reach visual illumination, the subsequent treatment of image is convenient in the atural object radiation recovering shadow region, needs to eliminate the shade on image.More common shadow removing method one utilizes remote sensing image to carry infrared band to carry out radiation compensation at present, and the general effectiveness comparison of this method is good, also relatively more accurate, but current most high-resolution remote sensing image is not with infrared band; Two is utilize the shade formation model of hypothesis to carry out radiometric restoration to shadow region, and this method cannot obtain desired result when video imaging environment does not meet hypothesized model, bad in the boundary treatment of details effect of shade in addition; Three is carry out integral radiation process to remote sensing image, make shadow region obtain radiation compensation, but this method is while compensating to shadow information, also the overall information changing non-hatched area, and compensation effect cannot be passed judgment on.
Summary of the invention
For the problems referred to above, the present invention proposes a kind of shadow removing method based on checking lines matching inside and outside shadow region, the method target in the visual effect macroscopically eliminating shade, microcosmic also will make shadow edge transition naturally leave no trace.
Technical scheme of the present invention is a kind of shadow removing method based on checking lines matching inside and outside shadow region, comprises the following steps:
Step 1, carries out shadow Detection to image, obtains the shadow region in image and corresponding shadow edge line of vector;
Step 2, if the corresponding shadow edge line of vector in the arbitrary shadow region of gained is R in step 1, utilizing shadow edge line of vector R to generate, border, shadow region is inside and outside checks line, border, described shadow region is inside and outside check line comprise extended out by shadow edge line of vector R after be positioned at the outer inspection line R of non-hatched area
1with inside contracted by shadow edge line of vector R after be positioned at shadow object and check line R
2,
Step 3, according to the outer inspection line R of arbitrary shadow region
1with interior inspection line R
2, respectively following steps are performed, step 3.1 to each wave band, respectively along outer inspection line R
1with interior inspection line R
2gather the gray-scale value of image on current band, generate grey scale curve V
1with grey scale curve V
2;
Step 3.2, to step 3.1 gained grey scale curve V
1with grey scale curve V
2mate, obtain grey scale curve V
1with grey scale curve V
2between radiation difference;
Step 3.3, whether determining step 3.2 gained radiation difference meet the iteration termination condition preset, if not then with the radiation characteristic of nonshaded area for reference, radiant correction process is carried out to shadow region, then returns step 3.1, if then enter step 3.4;
Step 3.4, checks line R in addition
1with interior inspection line R
2between as zone of transition, in zone of transition, emergence transition processing is carried out to the border of the shadow region after current radiation correction process.
And, shadow edge line of vector R is utilized to generate the inside and outside inspection in border, shadow region line in step 2, implementation is as follows, each some j on shadow edge line of vector R is got to the vertical line of shadow edge line of vector R, and on vertical line, both sides are that the point of a is respectively as outer inspection line R with some j distance
1with interior inspection line R
2the node of middle correspondence, outer inspection line R
1all nodes connect and compose outer inspection line R
1, interior inspection line R
2all nodes connect and compose interior inspection line R
2, wherein a is parameter preset.
And, generate grey scale curve V in step 3.1
1with grey scale curve V
2time, make grey scale curve V
1with grey scale curve V
2sampled pixel quantity is identical and position corresponding.
And, to step 3.1 gained grey scale curve V in step 3.2
1with grey scale curve V
2before mating, to grey scale curve V
1with grey scale curve V
2do Gaussian smoothing filter process.
And the implementation of step 3.2 is, to the grey scale curve V after Gaussian smoothing filter process
1with grey scale curve V
2with the average segmentation of same scale, then to gained sectional curve calculated curve shape similarity, get rid of some sectional curves that related coefficient is minimum; Calculate gray scale mean difference D according to remaining segment curve, obtain grey scale curve V
1with grey scale curve V
2between radiation difference.
And, the radiation characteristic of nonshaded area described in step 3.3, grey scale curve V after the some sectional curves adopting eliminating related coefficient minimum
1statistical property.
And the iteration termination condition preset described in step 3.3 is that this gained radiation difference is less than a certain predetermined threshold value T
2and the variable quantity of adjacent n iteration gained radiation difference is less than a certain predetermined threshold value T
1, n is preset times.
The present invention is to check inside and outside shadow region that the intensity distribution difference along the line of line represents external radiation difference in shadow region, irradiation treatment algorithm flow is simple, workable, shadow region recovery effects is more natural, the macroscopical difference can looking after shadow region and background is eliminated, and also can realize shadow edge place nuance and eliminate; Further, carry out shadow removing with the method for iterative radiometric calibration, computing method simple and stable, treatment effect is controlled.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the embodiment of the present invention.
Fig. 2 is the inside and outside inspection in border, the shadow region line schematic diagram of the embodiment of the present invention.
Fig. 3 is the inside and outside node correspondence schematic diagram checking line in the border, shadow region of the embodiment of the present invention.
Fig. 4 is the grey scale curve schematic diagram that the inside and outside inspection line of the embodiment of the present invention is corresponding.
Fig. 5 is result figure after the inside and outside inspection line corresponding grey scale curve Gaussian smoothing of the embodiment of the present invention.
Fig. 6 is the zone of transition schematic diagram of the embodiment of the present invention.
Fig. 7 is the raw video schematic diagram of the embodiment of the present invention.
Fig. 8 is the shadow removing result schematic diagram of the embodiment of the present invention.
Embodiment
Provided by the present invention a kind of based on checking inside and outside shadow region that the shadow removing method of lines matching is, shadow detection method is utilized to obtain shadow region, and generate shade inside and outside inspection line, then with the radiation characteristic of non-hatched area for benchmark, iteration irradiation treatment is carried out to recover atural object radiation characteristic in shadow region to shadow region, and with the luminance difference (radiation difference) along inspection line for calibration result check criteria, finally utilize inside and outside inspection line to generate transitional region, emergence process is carried out to transitional region.During concrete enforcement, computer software technology can be adopted to realize automatic flow and to run, describe technical solution of the present invention in detail below in conjunction with drawings and Examples.
As shown in Figure 1, the flow process of embodiment specifically comprises the following steps:
Step 1, carries out shadow Detection to image, obtains the shadow region in image and shadow edge line of vector.
Shadow region in image may have multiple, all performs same flow process process arbitrary shadow region.Idiographic flow can be designed to, initialization i=1, carries out following process to i-th shadow region in image.
Step 2, if the corresponding shadow edge line of vector in the arbitrary shadow region of gained is R in step 1, utilizes shadow edge line of vector R to generate the inside and outside inspection in border, shadow region line.
As Fig. 2, if the shadow edge line of vector of i-th shadow region is R in image, border, shadow region is inside and outside check line comprise extended out by shadow edge line of vector R after be positioned at the outer inspection line R of non-hatched area
1with inside contracted by shadow edge line of vector R after be positioned at shadow object and check line R
2.As Fig. 3, make outer inspection line R
1with interior inspection line R
2in each node mutually corresponding.Article one, shadow edge line of vector comprises several nodes.
Because inside and outside inspection line Length discrepancy, for making each node of inside and outside inspection line mutually corresponding, embodiment for reference, gets the vertical line of shadow edge line of vector R with shadow edge line of vector to each node j on shadow edge line of vector, on vertical line, both sides are that the point of a is respectively as outer inspection line R with some j distance
1with interior inspection line R
2the node of middle correspondence.Outer inspection line R
1all nodes connect and compose outer inspection line R
1, interior inspection line R
2all nodes connect and compose interior inspection line R
2.The node of inside and outside inspection line can walking of reference shadow boundary vector line always be determined, the such as clockwise words left side is outside and checks line, and the right is that inner side checks line.During concrete enforcement, the straight line gernertion of several matchings that the vertical line putting j top shadow boundary vector line can be closed on shadow edge line of vector by a j.Wherein a is parameter preset, can be specified voluntarily during concrete enforcement by user according to actual conditions.
Step 3, according to the outer inspection line R of arbitrary shadow region
1with interior inspection line R
2, respectively following steps are performed to each wave band,
According to the art custom, image is generally divided into RGB(red, green, blue) three wave bands.The present invention proposes to process respectively three wave bands, and flow process can be designed as and processes respectively successively (order does not require) three wave bands, can obtain the shadow removing result to raw video after processing.
Step 3.1, respectively along outer inspection line R
1with interior inspection line R
2gather the gray-scale value of corresponding image on current band, generate grey scale curve V
1with grey scale curve V
2.
As shown in Figure 4, along outer inspection line R
1gather the grey scale curve V that corresponding image band grey data generates
1be in top, along interior inspection line R
2gather the grey scale curve V that corresponding image band grey data generates
2be in below.
Generate grey scale curve V
1with grey scale curve V
2time, standardization is carried out to grey scale curve, namely make two grey scale curve sampled pixel (node) quantity identical, and spatial relationship is corresponding, is convenient to carry out comparison in difference and coupling.Because step 2 achieves outer inspection line R
1with interior inspection line R
2each node is mutually corresponding, the grey scale curve V that step 3.1 generates
1with grey scale curve V
2spatial relationship also can be corresponding.
Step 3.2, to grey scale curve V
1with grey scale curve V
2mate, obtain grey scale curve V
1with grey scale curve V
2between radiation difference.
In order to make the grey scale curve after processing more can represent radiation difference inside and outside shadow region, can to grey scale curve V before coupling
1with grey scale curve V
2carry out gaussian filtering smoothing processing, extract the low frequency part of two grey scale curve, the interference of local nuance can be eliminated like this, and the image radiation characteristic of two inspection line correspondence positions can be reflected.As shown in Figure 5, grey scale curve V
1gaussian smoothing after result be in top, grey scale curve V
2gaussian smoothing after result be in below.During concrete enforcement, the size of Gaussian smoothing template and filter times can be used as parameter and determine according to the smoothness of the feature of actual image and grey scale curve itself.
External radiation difference in the life shadow region that line corresponding grey scale curve reflects is checked inside and outside obtaining, can two grey scale curve with the average segmentation of same criterion (such as two the corresponding yardstick intervals of grey scale curve on transverse axis are equal), then following formula 1 calculated curve shape similarity is utilized to gained sectional curve, get rid of some sectional curves that related coefficient is minimum, make it not participate in subsequent calculations.As long as most of part can reflect that inside and outside shadow region, radiation characteristic just can carry out irradiation treatment on curve from probability, although therefore average segmentation algorithm is simple, reasonable result can be obtained at majority of case.If think the precision improving further coupling, can to the minimum sectional curve both sides adjacent sectional curve two stage cultivation again of related coefficient.Because the not corresponding part of more shape also may be there is in the both sides adjacent sectional of the sectional curve that related coefficient is little, if participate in subsequent calculations also have certain influence to precision, so respectively can to try again average segmentation to both sides adjacent sectional curve, further eliminating some sectional curves that wherein related coefficient is minimum, just as to original grey scale curve V
1, V
2do staging treating the same.
If grey scale curve V
1, V
2in any pair corresponding segment curve, comprise grey scale curve V
1in sectional curve A and grey scale curve V
2in sectional curve B, calculated curve shape similarity can adopt following formula
Wherein Shape (A, B) the curve shape similarity of sectional curve A and sectional curve B is represented, its essence is the related coefficient of sectional curve A and sectional curve B, n is the node number of sectional curve A and sectional curve B, and the node number of sectional curve A is identical with the node number of sectional curve B.
refer to the gray-scale value of upper i-th point of sectional curve A,
refer to the gray-scale value of upper i-th point of sectional curve B, the value of i is 1,2 ... n;
to refer on sectional curve A have the mean value of a gray-scale value,
to refer on sectional curve B the mean value of gray-scale value a little.
Calculate similarity process and be matching process, when Similarity measures result is greater than certain given threshold value, represent that upper and lower two sectional curve shapes are similar.After obtaining corresponding segments curve by coupling, utilize corresponding segments curve can obtain radiation difference between two grey scale curve, that expresses radiation characteristic and difference inside and outside shadow region mainly can adopt gray scale mean difference D, the present invention replaces carrying out parametric statistics, grey scale curve V inside and outside shadow region by two grey scale curve respectively
1with grey scale curve V
2between radiation difference adopt following formulae discovery:
According to prior art, gray scale mean difference D computing formula is as follows,
Wherein
grey scale curve V
1a middle jth gray-scale value,
grey scale curve V
2a middle jth gray-scale value, N is grey scale curve V
1with grey scale curve V
2interior joint number, grey scale curve V
1with grey scale curve V
2interior joint number is equal, and the value of j is 1,2 ... N.
The effect of this coupling acquisition two grey scale curve to reflect to greatest extent shadow region forms the part of the radiation characteristic of front and back.Because the atural object of boundary vector line both sides, shadow region radiation characteristic before and after shade is formed might not be identical, such as the buildings of shade meeting and formation shade is adjacent, understand the highlighted radiation characteristic that some is buildings in the grey scale curve that outer inspection line is formed, the radiation characteristic of this partial radiation characteristic and interior inspection line correspondence position does not have comparability.Step 3.3, whether determining step 3.2 gained radiation difference meet the iteration termination condition preset, if not then with the radiation characteristic of nonshaded area for reference, radiant correction process is carried out to shadow region, then returns step 3.1, if then enter step 3.4.When not meeting the iteration termination condition preset, return step 3.1 and carry out next round iteration, comprise the image based on epicycle gained after radiant correction process is carried out in step 3.3 pair shadow region, again from respectively along outer inspection line R
1with interior inspection line R
2gather corresponding image current band gray-scale value, generate new grey scale curve V
1with grey scale curve V
2, then same execution processes until meet the iteration termination condition preset.
First embodiment judges whether gained radiation difference meets the iteration termination condition preset, whether the parameter value of i.e. radiation difference is enough little, if not then with the radiation characteristic of non-hatched area for benchmark, atural object radiation characteristic in radiation intensification process recovery shadow region is carried out to shadow region, makes grey scale curve V by successive ignition
1with grey scale curve V
2between radiation difference minimum and in threshold range.
Be reference by surrounding's non-hatched area of shadow region, the shadow region like this after radiant correction (shadow removing) is just not easy to be discovered, and can obtain desirable effect.For the purpose of raising the efficiency, what the radiation characteristic of the non-hatched area of reference of the present invention can be similar to is substituted by the outer inspection line generated.What the radiation characteristic of the non-hatched area of embodiment adopted is that epicycle externally checks line R
1after removing the little sectional curve of wherein related coefficient, the statistical property of the gray scale of the corresponding wave band on outer inspection line, i.e. grey scale curve V
1the statistical property of remaining segment curve.
Therefore, in embodiment, the irradiation treatment of shadow region is an iterative process, checks that the generation of line corresponding grey scale curve, Gaussian smoothing, coupling, radiation difference get the several committed step of shadow region irradiation treatment inside and outside comprising.Take non-hatched area as reference, carry out shadow removing with relative detector calibration method, after each radiant correction, again add up grey scale curve V
1with grey scale curve V
2radiation difference, stop when meeting iteration termination condition.If the variable quantity of radiation discrepancy delta is less than a certain value T after can being set as the iterative processing of adjacent n time
1and current gained radiation difference is less than a certain threshold value T
2, then think that radiation discrepancy delta has reached minimum, stop iterative radiometric calibration carry out step 3.4.T
1and T
2be all the threshold value of setting, during concrete enforcement, those skilled in the art can according to the different value of the different set of image; N is preset times, and during concrete enforcement, those skilled in the art can need according to precision and experiment value.If be xth time iteration current, iteration termination condition is: the radiation difference that xth time performs gained is less than T
2and the radiation difference that the radiation difference of xth time execution gained and xth perform gained for-1 time subtracts each other the variable quantity delta (1) obtained, the difference of the difference of xth time and xth-2 times subtracts each other the variable quantity delta (2) obtained ... the difference that the difference of xth time is secondary with xth-n subtracts each other variable quantity delta (n) obtained, this n variable quantity delta (1), delta (2) ... delta (n) is less than T
1.
According to the radiation difference obtained, take nonshaded area as reference, carry out radiant correction process to shadow region, concrete radiation correction method can be selected based on relative radiometric correction methods such as maximin method, histogram matching, mist elimination correction method, mean variance method or simple linear regression methods.Such as grey scale curve V
1with grey scale curve V
2between radiation difference can represent by the gray scale mean difference D between use curve, be then applicable to using maximin method to carry out irradiation treatment, be implemented as prior art, it will not go into details in the present invention.
Step 3.4, checks line R in addition
1with interior inspection line R
2between as zone of transition, in zone of transition, emergence transition processing is carried out to the border of the shadow region after current radiation correction process.
Aftertreatment is carried out to the result (shadow region namely after the process of the last execution step 3.3 radiant correction) completing shadow region irradiation treatment, sets up zone of transition with inside and outside inspection line, emergence process is carried out to shadow edge place; With check line for benchmark generate zone of transition as shown in Figure 6, utilize emergence transition processing can obtain shadow removing result, as Fig. 8, be implemented as prior art, it will not go into details in the present invention.Comparison diagram 7, visible technique effect of the present invention.
Idiographic flow can be designed to process the shadow region in image successively, if when pre-treatment is i-th shadow region, initialization i=1, to current shadow region perform step 2,3 process complete after, make i=i+1, then return step 2 and continue to process next shadow region, until process all shadow regions in image, namely i=M, M are the shadow region sum in image.The shadow removing result to view picture image can be obtained like this.
Specific embodiment described herein is only to the explanation for example of the present invention's spirit.Those skilled in the art can make various amendment or supplement or adopt similar mode to substitute to described specific embodiment, but can't depart from spirit of the present invention or surmount the scope that appended claims defines.
Claims (4)
1., based on the shadow removing method checking lines matching inside and outside shadow region, it is characterized in that, comprise the following steps:
Step 1, carries out shadow Detection to image, obtains the shadow region in image and corresponding shadow edge line of vector;
Step 2, if the corresponding shadow edge line of vector in the arbitrary shadow region of gained is R in step 1, utilizing shadow edge line of vector R to generate, border, shadow region is inside and outside checks line, border, described shadow region is inside and outside check line comprise extended out by shadow edge line of vector R after be positioned at the outer inspection line R of non-hatched area
1with inside contracted by shadow edge line of vector R after be positioned at shadow object and check line R
2,
Step 3, according to the outer inspection line R of arbitrary shadow region
1with interior inspection line R
2, respectively following steps are performed to each wave band,
Step 3.1, respectively along outer inspection line R
1with interior inspection line R
2gather the gray-scale value of image on current band, generate grey scale curve V
1with grey scale curve V
2, make grey scale curve V
1with grey scale curve V
2sampled pixel quantity is identical and position corresponding;
Step 3.2, to step 3.1 gained grey scale curve V
1with grey scale curve V
2mate, obtain grey scale curve V
1with grey scale curve V
2between radiation difference; Wherein, to step 3.1 gained grey scale curve V
1with grey scale curve V
2before mating, to grey scale curve V
1with grey scale curve V
2do Gaussian smoothing filter process, coupling implementation is, to the grey scale curve V after Gaussian smoothing filter process
1with grey scale curve V
2with the average segmentation of same scale, then to gained sectional curve calculated curve shape similarity, get rid of some sectional curves that related coefficient is minimum; Calculate gray scale mean difference D according to remaining segment curve, obtain grey scale curve V
1with grey scale curve V
2between radiation difference;
Step 3.3, whether determining step 3.2 gained radiation difference meet the iteration termination condition preset, if not then with the radiation characteristic of nonshaded area for reference, radiant correction process is carried out to shadow region, then returns step 3.1, if then enter step 3.4;
Step 3.4, checks line R in addition
1with interior inspection line R
2between as zone of transition, in zone of transition, emergence transition processing is carried out to the border of the shadow region after current radiation correction process.
2. according to claim 1 based on the shadow removing method checking lines matching inside and outside shadow region, it is characterized in that: utilize shadow edge line of vector R to generate the inside and outside inspection in border, shadow region line in step 2, implementation is as follows,
Each some j on shadow edge line of vector R is got to the vertical line of shadow edge line of vector R, on vertical line, both sides are that the point of a is respectively as outer inspection line R with some j distance
1with interior inspection line R
2the node of middle correspondence, outer inspection line R
1all nodes connect and compose outer inspection line R
1, interior inspection line R
2all nodes connect and compose interior inspection line R
2, wherein a is parameter preset.
3. according to claim 2 based on the shadow removing method checking lines matching inside and outside shadow region, it is characterized in that: the radiation characteristic of nonshaded area described in step 3.3, grey scale curve V after the some sectional curves adopting eliminating related coefficient minimum
1statistical property.
4. according to claim 1 or 2 or 3 based on the shadow removing method checking lines matching inside and outside shadow region, it is characterized in that: described in step 3.3 preset iteration termination condition be that this gained radiation difference is less than a certain predetermined threshold value T
2and the variable quantity of adjacent n iteration gained radiation difference is less than a certain predetermined threshold value T
1, n is preset times.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210342195.9A CN102915524B (en) | 2012-09-14 | 2012-09-14 | Method for eliminating shadow based on match of inside and outside check lines of shadow area |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210342195.9A CN102915524B (en) | 2012-09-14 | 2012-09-14 | Method for eliminating shadow based on match of inside and outside check lines of shadow area |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102915524A CN102915524A (en) | 2013-02-06 |
CN102915524B true CN102915524B (en) | 2015-01-14 |
Family
ID=47613879
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210342195.9A Active CN102915524B (en) | 2012-09-14 | 2012-09-14 | Method for eliminating shadow based on match of inside and outside check lines of shadow area |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102915524B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113240619B (en) * | 2021-01-26 | 2022-06-21 | 重庆师范大学 | Soil image shadow detection method based on improved subtraction histogram |
CN113487502B (en) * | 2021-06-30 | 2022-05-03 | 中南大学 | Shadow removing method for hollow image |
CN116109540B (en) * | 2023-03-22 | 2023-07-18 | 智洋创新科技股份有限公司 | Image registration fusion method and system based on particle swarm optimization gray curve matching |
-
2012
- 2012-09-14 CN CN201210342195.9A patent/CN102915524B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN102915524A (en) | 2013-02-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhang et al. | Object-oriented shadow detection and removal from urban high-resolution remote sensing images | |
CN108389188B (en) | Sparse hyperspectral abnormal target detection method | |
CN109615653B (en) | Leakage water area detection and identification method based on deep learning and visual field projection model | |
CN103971115B (en) | Automatic extraction method for newly-increased construction land image spots based on NDVI and PanTex index | |
CN101266645B (en) | Iris positioning method based on multi-resolution analysis | |
CN100373397C (en) | Pre-processing method for iris image | |
CN105354865A (en) | Automatic cloud detection method and system for multi-spectral remote sensing satellite image | |
CN109146948A (en) | The quantization of crop growing state phenotypic parameter and the correlation with yield analysis method of view-based access control model | |
CN102254319A (en) | Method for carrying out change detection on multi-level segmented remote sensing image | |
CN112001868A (en) | Infrared and visible light image fusion method and system based on generation of antagonistic network | |
CN103679714A (en) | Method for automatic registration of optical image and SAR image based on gradient cross-correlation | |
CN113239830B (en) | Remote sensing image cloud detection method based on full-scale feature fusion | |
Raffei et al. | A low lighting or contrast ratio visible iris recognition using iso-contrast limited adaptive histogram equalization | |
CN103413120A (en) | Tracking method based on integral and partial recognition of object | |
CN105809140A (en) | Method and device for extracting surface water body information based on remote sensing model | |
CN104732492A (en) | Depth image denoising method | |
CN110059640A (en) | The in-orbit recognition methods of sea ship based on Optical remote satellite near-infrared spectral coverage | |
CN104021559A (en) | Image registration method based on mutual information and Harris corner point detection | |
CN102915524B (en) | Method for eliminating shadow based on match of inside and outside check lines of shadow area | |
CN107992856B (en) | High-resolution remote sensing building shadow detection method under urban scene | |
CN107564017A (en) | A kind of city high score remote sensing image shadow Detection and dividing method | |
CN106778629A (en) | Greenhouse recognition methods and device | |
CN104197933A (en) | Method for enhancing and extracting high stars and other sliding fixed stars in view field of telescope | |
US11151377B2 (en) | Cloud detection method based on landsat 8 snow-containing image | |
CN103116747A (en) | Method and system for automatically recognizing images of stalks and leaves of corns |
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 |