KR101681178B1 - Satellite image processing method and apparatus - Google Patents
Satellite image processing method and apparatus Download PDFInfo
- Publication number
- KR101681178B1 KR101681178B1 KR1020150169802A KR20150169802A KR101681178B1 KR 101681178 B1 KR101681178 B1 KR 101681178B1 KR 1020150169802 A KR1020150169802 A KR 1020150169802A KR 20150169802 A KR20150169802 A KR 20150169802A KR 101681178 B1 KR101681178 B1 KR 101681178B1
- Authority
- KR
- South Korea
- Prior art keywords
- cloud
- local
- satellite image
- region
- area
- Prior art date
Links
Images
Classifications
-
- G06T7/0081—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a satellite image processing method and apparatus capable of detecting and dividing a cloud in a satellite image. The satellite image processing method includes receiving satellite images, determining whether clouds exist in the satellite images, A step of dividing a region including a pixel having a brightness of a predetermined threshold value or more into a cloud center region in a satellite image when it is determined that a cloud exists in the satellite image, And applying a local window to the extended uncertainty region to divide it into cloud edge regions. According to the present invention, it is possible to detect and divide a cloud very accurately in a satellite image.
Description
The present invention relates to a satellite image processing method and apparatus, and more particularly, to a satellite image processing method and apparatus capable of detecting and dividing a cloud in a satellite image.
Satellite information utilization is important for national infrastructural technology such as establishment of management and operation plan of various facilities and infrastructure, urban planning and establishment of national management policy. In addition, it has high added value that can make a great contribution to utilization of domestic and overseas image information and export. Ideally, it is desirable to utilize all the information that can be obtained from the satellite.
However, according to the Ice, Cloud, and land Elevation Satellite (ICESat), about 70% of the earth is always covered with clouds. In addition, KOMSAT-1 showed about 25% effective image acquisition rate of about 25%, excluding the images containing clouds among about 260,000 collected images.
Figure 1 shows a satellite image including clouds.
Referring to FIG. 1, a satellite image including a cloud may be divided into a thick cloud, a thin cloud, and a ground.
Previously, cloud detection in satellite images meant dividing the center of the cloud, which roughly included the edge of the cloud, rather than accurately detecting the edge of the cloud. In the conventional method, a method using a pixel value and a threshold value of an image is mainly used.
2 is a satellite image photograph provided to explain a method of detecting a cloud in a conventional satellite image.
Referring to FIG. 2, it is possible to divide only the approximate shape of the cloud, including a part of the edge of the cloud and the center of the cloud, without dividing the edge of the cloud accurately by a conventionally known method.
As a result, there is a problem that satellite image information should be utilized in a state in which a large portion of the cloud edge is not removed from the satellite image.
SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a satellite image processing method and apparatus capable of accurately detecting and dividing a cloud from a satellite image to a cloud edge.
According to an aspect of the present invention, there is provided a satellite image processing method including receiving a satellite image, determining whether a cloud exists in the satellite image, A cloud dividing step of dividing an area made up of pixels having a brightness of a predetermined threshold value or more in the satellite image into a cloud center area and a local window in the uncertainty area extended from the cloud center area, And a cloud segmentation step by a local method for dividing the cloud segment region into a cloud edge region.
The method may further comprise removing an area having a predetermined shape or area from the area divided into the cloud center area and the cloud edge area in the satellite image in the area divided by the clouds.
Wherein the step of determining whether a cloud exists in the satellite image comprises the steps of obtaining an average brightness value of the satellite image as an initial threshold value and a first region of pixels having a brightness value larger than the average brightness value of the satellite image, A second region including pixels having brightness values smaller than a brightness value, calculating a mean brightness value of the first region and an average brightness value of the second region as a correction threshold value, And determining that a cloud exists in the satellite image if the satellite image is larger than a predetermined reference.
The predetermined threshold value may be the correction threshold value.
The cloud center area consisting of pixels having a brightness value equal to or greater than the correction threshold value can be divided in the satellite image.
Wherein the step of dividing the cloud by the local method comprises the steps of: obtaining an extended uncertainty region from the cloud center region; and applying a local window having a predetermined size to each pixel belonging to the uncertainty region, And determining whether each pixel is a cloud or an earth surface.
Determining whether a pixel belonging to the uncertainty region is a cloud or a surface of a ground by applying the local window may include determining a minimum brightness value (Umax) among brightness values of pixels included in the local window and a minimum brightness value brightness value (Umin) the difference between local luminance difference (C local) to obtain, is greater than or equal to the local luminance difference (C local) the value (T ") set by the user, the area a (Umax + Umin) / 2 (T local ), and if the local luminance difference (C local ) is less than a value (T ") set by the user, the correction threshold value is set as a local threshold value (T local ) The center pixel of the region window is determined as a cloud and the brightness value Pcenter of the center pixel of the region window is less than the local threshold value T local if the brightness value Pcenter is greater than or equal to the local threshold value T local . The center pixel of the local window can be determined as the ground surface.
By applying morphological operator dilation to the satellite image, an uncertainty region extending from the cloud center region can be obtained.
According to an aspect of the present invention, there is provided a satellite image processing apparatus comprising: a cloud presence determination unit that receives a satellite image to determine whether a cloud exists in the satellite image; A global dividing unit that divides an area made up of pixels having a brightness of a predetermined threshold value or more in the satellite image into a cloud center area and a local window for the uncertainty area extended from the cloud center area, And includes a regional partition that divides into cloud edge regions.
According to the present invention, it is possible to detect and divide a cloud very accurately in a satellite image.
Figure 1 shows a satellite image including clouds.
2 is a satellite image photograph provided to explain a method of detecting a cloud in a conventional satellite image.
3 is a block diagram illustrating a configuration of a satellite image processing apparatus according to an embodiment of the present invention.
4 is a flowchart illustrating an operation of the satellite image processing apparatus according to an embodiment of the present invention.
5 is a flowchart illustrating a method for determining whether or not a cloud exists in a satellite image.
FIG. 6 is a flowchart illustrating a method of dividing a cloud edge region in a satellite image according to the present invention.
Figure 7 is a view provided to illustrate the uncertainty zone between the cloud center and the ground to which the regional method according to the present invention is applied.
FIG. 8 is a diagram provided to compare the result of dividing a cloud in a satellite image by the method according to the present invention and the conventional method.
3 is a block diagram illustrating a configuration of a satellite image processing apparatus according to an embodiment of the present invention.
3, the satellite
The cloud
When it is determined that the cloud does not exist in the satellite image, the cloud
The
The
The
The operation of the satellite image processing apparatus according to the present invention will now be described in more detail with reference to FIG.
4 is a flowchart illustrating an operation of the satellite image processing apparatus according to an embodiment of the present invention.
Referring to FIG. 4, first, when the satellite image is input (S410), the cloud presence determiner 110 determines whether there is a cloud in the input satellite image (S420).
In step S220, the cloud presence determiner 110 calculates an average brightness value of the inputted satellite image and sets it as an initial threshold value. Then, the cloud
Step S420 will be described in more detail with reference to FIG.
5 is a flowchart illustrating a method for determining whether or not a cloud exists in a satellite image.
First, the cloud
Where I (x, y) is the brightness value of the (x, y) pixel of the satellite image.
Next, the cloud
Here, P c is a region having a larger brightness value than the initial threshold value, and P g is a region having a brightness value less than the initial threshold value. Of course, according to the embodiment, it is also possible to divide a pixel having the same brightness value as the initial threshold value by including it in the Pc area.
Next, the cloud
Then, the cloud
Next, when the correction threshold value T 'is greater than a predetermined reference, the cloud presence determiner 110 determines that a cloud exists in the input satellite image, and in the opposite case, the cloud presence determiner 110 can determine that there is no cloud (S429). Here, the brightness value 500 may be used as a criterion for determining the presence of a cloud, but is not limited thereto and can be adjusted according to the embodiment.
Referring to FIG. 4 again, if it is determined that a cloud is present in the satellite image input by the cloud presence determination unit 110 (S420-Y), the
Thereafter, the
Step S440 will be described in more detail with reference to FIG.
6 is a flow chart for explaining a method of dividing a cloud edge portion by a local method in a satellite image according to the present invention.
Referring to FIG. 6, in step S430, the
Figure 7 is a view provided to illustrate the uncertainty zone between the cloud center and the ground to which the regional method according to the present invention is applied.
Referring to FIG. 7, the
Next, the
Here, Umax is the maximum brightness value among the brightness values of the pixels included in the local window, and Umin is the minimum brightness value among the brightness values of the pixels included in the local window.
T "may be appropriately set according to the embodiment with a value set by the user.
First, the difference between the maximum brightness value Umax among the brightness values of the pixels included in the local window and the minimum brightness value Umin among the brightness values of the pixels included in the local window is obtained as the local brightness difference C local (S4431)
If the local contrast (C local ) is greater than or equal to the predetermined T ", the local threshold T local can be set to (Umax + Umin) / 2. Conversely, if C local is less than T" , And the local threshold value (T local ) can be defined as a correction threshold value T 'obtained by a global method (S4433).
Next, if the brightness value Pcenter of the local window center pixel is equal to or greater than the local threshold value T local , the local window center pixel can be determined as a cloud. Conversely, when the brightness value Pcenter of the local window center pixel is smaller than the local threshold value T local , the local window center pixel may be determined as the ground surface (S4435).
In this way, the edge portion of the cloud can be divided by a local method by repeating the application of the local window having each pixel included in the uncertainty region as the center pixel to the input satellite image.
Referring again to FIG. 4, finally, the
FIG. 8 is a diagram provided to compare the result of dividing a cloud in a satellite image by the method according to the present invention and the conventional method.
8 (a) is an input satellite image, Figs. 8 (b) to 8 (g) are images obtained by dividing a cloud in a satellite image by a conventional method, and Fig. 8 , Which is an image obtained by dividing a cloud from a satellite image. It can be seen that the method according to the present invention is able to more accurately detect the cloud in the satellite image than the conventional method.
100: satellite image processing device
110: cloud existence discrimination unit
120: Global division
130: Regional division
140: Outlier removal
Claims (14)
Determining whether a cloud exists in the satellite image,
A cloud segmentation step by a global method of dividing an area made up of pixels having a brightness of a predetermined threshold value or more in the satellite image into a cloud center area when it is determined that a cloud exists in the satellite image,
A cloud splitting step by a local method of splitting an uncertainty area extended from the cloud center area into a cloud edge area by applying a local window, and
Removing a region having a predetermined shape or area from a region divided into a cloud center region and a cloud edge region in the satellite image from a region divided into clouds
And a satellite image processing method.
Determining whether a cloud exists in the satellite image,
A cloud dividing step by a global method of dividing an area made up of pixels having a brightness of a predetermined threshold value or more in the satellite image into a cloud center area when it is determined that a cloud exists in the satellite image,
A cloud partitioning step by a local method of dividing the uncertainty area extending from the cloud center area into a cloud edge area by applying a local window
Lt; / RTI >
Wherein the step of determining whether a cloud exists in the satellite image comprises:
Obtaining an average brightness value of the satellite image as an initial threshold value,
Dividing a first area made up of pixels having a brightness value greater than an average brightness value of the satellite image and a second area having a brightness value less than the average brightness value,
Calculating an average brightness value of the first area and an average brightness value of the second area as a correction threshold value, and
Determining that a cloud exists in the satellite image if the correction threshold value is greater than a predetermined criterion
And a satellite image processing method.
The predetermined threshold value is the correction threshold value,
Wherein the center of the cloud is divided into pixels having brightness values greater than or equal to the correction threshold value in the satellite image.
Wherein the step of dividing the cloud by the regional method comprises:
Obtaining an extended uncertainty region from the cloud center region, and
Determining whether each pixel belonging to the uncertainty region is a cloud or an earth surface by applying a local window having a predetermined size using each pixel as a center pixel;
And a satellite image processing method.
Determining whether a pixel belonging to the uncertainty region is a cloud or an earth surface by applying the region window,
A local brightness difference (C local ) is calculated as a difference between a maximum brightness value (Umax) and a minimum brightness value (Umin) among the brightness values of the pixels included in the local window,
If the local luminance difference C local is greater than or equal to the value T "set by the user, (U max + Umin) / 2 is set as the local threshold T local ,
If the local luminance difference (C local ) is less than the value (T ") set by the user, the correction threshold value is set as the local threshold (T local )
Determining a center pixel of the region window as a cloud if the brightness value of the center pixel of the region window is greater than or equal to the local threshold value T local and determining that the brightness value of the center pixel of the region window is Pcenter And determining the center pixel of the local window as the ground surface if the local window is smaller than the local threshold value (T local ).
And applying a morphological operator dilation to the satellite image to obtain an extended uncertainty region from the cloud center region.
A global segmentation unit for segmenting an area of a pixel having a brightness of a predetermined threshold value or more in the satellite image into a cloud center area when it is determined that a cloud exists in the satellite image;
A local division for dividing the uncertainty region extending from the cloud center region into a cloud edge region by applying a local window, and
And removing an area having a predetermined shape or area from the area divided into the cloud center area and the cloud edge area in the satellite image from the area divided by the cloud,
And a satellite image processing unit.
A global division unit for dividing an area including pixels having a brightness of a predetermined threshold value or more in the satellite image into a cloud center area when it is determined that a cloud exists in the satellite image,
A region dividing unit for dividing the uncertainty region extending from the cloud center region into a cloud edge region by applying a local window,
Lt; / RTI >
The cloud existence determining unit may determine,
And a second region including pixels having a brightness value smaller than the average brightness value, the first region being a pixel having a brightness value greater than the average brightness value of the satellite image and an average brightness value of the satellite image being an initial threshold value, And determining an average value of the average brightness value of the first area and the average brightness value of the second area as a correction threshold value and determining that a cloud exists in the satellite image when the correction threshold value is greater than a predetermined reference value Gt;
The predetermined threshold value is the correction threshold value,
Wherein the global division unit comprises:
Wherein the satellite image processor divides the cloud center area formed by pixels having a brightness value equal to or greater than the correction threshold value.
Wherein the local division unit comprises:
A satellite for determining an uncertainty region extending from the cloud center region and determining whether each pixel belongs to the uncertainty region is a cloud or an earth surface by applying a local window having a predetermined size using each pixel as a center pixel, Image processing apparatus.
Determining whether a pixel belonging to the uncertainty region is a cloud or an earth surface by applying the region window,
A local brightness difference (C local ) is calculated as a difference between a maximum brightness value (Umax) and a minimum brightness value (Umin) among the brightness values of the pixels included in the local window,
If the local luminance difference C local is greater than or equal to the value T "set by the user, (U max + Umin) / 2 is set as the local threshold T local ,
If the local luminance difference (C local ) is less than the value (T ") set by the user, the correction threshold value is set as the local threshold (T local )
Determining a center pixel of the region window as a cloud if the brightness value of the center pixel of the region window is greater than or equal to the local threshold value T local and determining that the brightness value of the center pixel of the region window is Pcenter And determines the central pixel of the local window to be the ground surface if the local window is smaller than the local threshold value (T local ).
Wherein the local division unit comprises:
And applying a morphological operator dilation to the satellite image to obtain an extended uncertainty region from the cloud center region.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150169802A KR101681178B1 (en) | 2015-12-01 | 2015-12-01 | Satellite image processing method and apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150169802A KR101681178B1 (en) | 2015-12-01 | 2015-12-01 | Satellite image processing method and apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
KR101681178B1 true KR101681178B1 (en) | 2016-11-30 |
Family
ID=57707399
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020150169802A KR101681178B1 (en) | 2015-12-01 | 2015-12-01 | Satellite image processing method and apparatus |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR101681178B1 (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101922645B1 (en) | 2017-06-09 | 2018-11-27 | 한국항공우주연구원 | cloud area detection device and cloud area detection method |
WO2021041918A1 (en) * | 2019-08-30 | 2021-03-04 | Numerica Corporation | System and method for space object detection in daytime sky images |
KR20220076079A (en) * | 2020-11-30 | 2022-06-08 | (주) 지오씨엔아이 | Map sheet image generation system and method using satellite images |
US11620819B2 (en) | 2021-08-27 | 2023-04-04 | Si Analytics Co., Ltd | Method for scheduling of shooting satellite images based on deep learning |
CN116862942A (en) * | 2023-09-05 | 2023-10-10 | 青岛哈尔滨工程大学创新发展中心 | Sea Wen Fanyan precision correction method based on cloud detection cloud removal and angle correction |
-
2015
- 2015-12-01 KR KR1020150169802A patent/KR101681178B1/en active IP Right Grant
Non-Patent Citations (1)
Title |
---|
영상에서의 구름 검출 및 제거를 위한 알고리즘 |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101922645B1 (en) | 2017-06-09 | 2018-11-27 | 한국항공우주연구원 | cloud area detection device and cloud area detection method |
WO2021041918A1 (en) * | 2019-08-30 | 2021-03-04 | Numerica Corporation | System and method for space object detection in daytime sky images |
US11587311B2 (en) | 2019-08-30 | 2023-02-21 | Slingshot Aerospace, Inc. | System and method for space object detection in daytime sky images |
KR20220076079A (en) * | 2020-11-30 | 2022-06-08 | (주) 지오씨엔아이 | Map sheet image generation system and method using satellite images |
KR102428049B1 (en) | 2020-11-30 | 2022-08-03 | (주) 지오씨엔아이 | Map sheet image generation system and method using satellite images |
US11620819B2 (en) | 2021-08-27 | 2023-04-04 | Si Analytics Co., Ltd | Method for scheduling of shooting satellite images based on deep learning |
CN116862942A (en) * | 2023-09-05 | 2023-10-10 | 青岛哈尔滨工程大学创新发展中心 | Sea Wen Fanyan precision correction method based on cloud detection cloud removal and angle correction |
CN116862942B (en) * | 2023-09-05 | 2023-12-19 | 青岛哈尔滨工程大学创新发展中心 | Sea Wen Fanyan precision correction method based on cloud detection cloud removal and angle correction |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101681178B1 (en) | Satellite image processing method and apparatus | |
CN108986119B (en) | Image segmentation method and device, computer equipment and readable storage medium | |
CN108549874B (en) | Target detection method, target detection equipment and computer-readable storage medium | |
CN106683052B (en) | Defogging photographs and videos using visual artifact suppression | |
CN105608455B (en) | A kind of license plate sloped correcting method and device | |
KR101717613B1 (en) | The moving vehicle detection system using an object tracking algorithm based on edge information, and method thereof | |
CN103679733B (en) | A kind of signal lamp image processing method and its device | |
WO2012074361A1 (en) | Method of image segmentation using intensity and depth information | |
JP2012038318A (en) | Target detection method and device | |
CN103500451B (en) | A kind of independent floating ice extracting method for satellite data | |
CN104778723A (en) | Method for performing motion detection on infrared image with three-frame difference method | |
JP6771134B2 (en) | Image correction method and image correction device | |
US20170358089A1 (en) | Object identifying apparatus | |
CN113628202B (en) | Determination method, cleaning robot and computer storage medium | |
Sun et al. | Fast single image dehazing using iterative bilateral filter | |
WO2019001698A1 (en) | Advanced driver assistance system and method | |
CN103971347A (en) | Method and device for treating shadow in video image | |
Othman et al. | Enhanced single image dehazing technique based on hsv color space | |
He et al. | Effective haze removal under mixed domain and retract neighborhood | |
KR20130124659A (en) | System for detaching object and the method thereof | |
CN116310889A (en) | Unmanned aerial vehicle environment perception data processing method, control terminal and storage medium | |
JP4054125B2 (en) | Moving image object extraction device | |
JP2024060851A (en) | Aerial image change extraction device | |
KR101553140B1 (en) | Method for compensating fogs image based on segmention | |
Kawarabuki et al. | Snowfall detection in a foggy scene |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
E902 | Notification of reason for refusal | ||
E701 | Decision to grant or registration of patent right | ||
GRNT | Written decision to grant | ||
FPAY | Annual fee payment |
Payment date: 20191016 Year of fee payment: 4 |