CN108280810B - Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image - Google Patents
Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image Download PDFInfo
- Publication number
- CN108280810B CN108280810B CN201810018545.3A CN201810018545A CN108280810B CN 108280810 B CN108280810 B CN 108280810B CN 201810018545 A CN201810018545 A CN 201810018545A CN 108280810 B CN108280810 B CN 108280810B
- Authority
- CN
- China
- Prior art keywords
- coverage area
- cloud coverage
- area
- cloud
- thick
- 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
- 230000003287 optical effect Effects 0.000 title claims abstract description 24
- 238000003672 processing method Methods 0.000 title claims abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 38
- 230000008439 repair process Effects 0.000 claims abstract description 20
- 239000011159 matrix material Substances 0.000 claims abstract description 19
- 238000000605 extraction Methods 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 12
- 238000009499 grossing Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000000007 visual effect Effects 0.000 claims description 5
- 230000007704 transition Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 230000000644 propagated effect Effects 0.000 claims 1
- 238000001914 filtration Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000010191 image analysis Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The patent provides an automatic processing method for repairing a cloud coverage area of a single-time phase optical remote sensing image, which aims at the limitation of a traditional cloud coverage area image repairing method in an optical remote sensing image. Firstly, in a thick cloud coverage area extraction stage, providing an extraction method based on color and texture characteristics, and screening a thick cloud coverage area by utilizing RGB color space and co-occurrence matrix contrast; secondly, in a cloud coverage area repairing stage, an improved Criminisi algorithm is provided for repairing a thick cloud coverage area and a thin cloud coverage area. According to the method, the cloud coverage area can be efficiently repaired without multi-temporal data, historical data and multi-source data, and the rapid and reliable emergency application of the optical remote sensing image in the repair of the cloud coverage area is realized.
Description
Technical Field
The invention relates to a processing method of a remote sensing image, in particular to an effective repairing and visual improving method for a cloud coverage area in a single-time phase optical remote sensing image.
Background
The optical remote sensing image is an important data source in the remote sensing application fields of land utilization, weather, environmental monitoring and the like. However, the optical image is easily affected by cloud and fog, which brings great difficulty to the subsequent analysis and interpretation of the image, such as identification and classification of ground objects in the remote sensing image; in addition, the visibility of the remote sensing image is greatly influenced. Therefore, the method has very important practical significance in removing or weakening the influence of the cloud layer in the application of the remote sensing image and improving the usability and interpretability of the remote sensing data by using the image processing technology. The traditional remote sensing image cloud removing and repairing method can be mainly divided into three categories: based on prior model, based on different source remote sensing data in the same region, and based on the same source multi-temporal remote sensing data. The cloud coverage area repairing method based on the prior model mostly depends on characteristic sparse dictionary establishment for a large number of remote sensing images, the established sparse dictionary is used for repairing cloud coverage area default data in a repairing stage, such as northwest industrial university and li zhang, the sparse dictionary obtained through characteristic block learning is used for repairing the cloud coverage area, and the method needs a large number of historical remote sensing data for modeling. Most of cloud coverage area repairing methods based on different source remote sensing data in the same area reconstruct image information of a cloud coverage area by using information of different wave bands in multispectral images and infrared images in the same area. For example, in the university of electronic technology, royal courage, etc., the processed result after cloud removal is restored through the remote sensing images of visible light and near infrared bands. A cloud coverage area repairing method based on homologous multi-temporal remote sensing data mainly predicts a cloud coverage area of a current image by using cloud coverage area-free images in different time phases. For example, tianjin han ming technologies development limited, yoyanhua, etc., repairs the cloud coverage area by strictly registering two images with and without clouds at different time phases.
In summary, in terms of repairing the cloud coverage area of the optical remote sensing image, the existing methods mostly require multi-temporal data of the same area, data of different sources in the same area, or prior information of historical data. With the increase of the occurrence frequency of various emergency incident situations, the method needs support of multiple time phases, multiple sources or historical data. However, these conditions are not met in many emergency remote sensing data applications. Therefore, a method for directly and effectively repairing a cloud coverage area based on single-time-phase remote sensing image analysis and processing is urgently needed.
Disclosure of Invention
Aiming at the limitation of the traditional remote sensing image cloud coverage area repairing method, the invention provides an automatic processing method for repairing a single-time phase optical remote sensing image cloud coverage area, and at present, no report of the method exists in China.
The method provided by the invention comprises two main steps of extracting a thick cloud coverage area based on color and texture characteristics and repairing the cloud coverage area based on an improved Criminisi algorithm, and specifically comprises the following steps:
first step is color and texture feature based thick cloud coverage area extraction
Aiming at the characteristic that the color and the textural features of a thick cloud coverage area of a large-view-field optical remote sensing image have obvious differences relative to other ground objects, the method firstly utilizes an RGB color space to carry out primary screening on a suspected thick cloud coverage area based on the color features; and then, determining candidate cloud coverage areas based on the texture features for the thick cloud coverage areas by using the gray level co-occurrence matrix contrast parameter.
Second step is based on improving cloud coverage area restoration of Criminisi algorithm
Based on the thick cloud coverage area screened in the last step, the improved Criminisi algorithm is adopted to repair the cloud coverage area. The method comprises the following specific steps: the method comprises the steps of determining the removal priority of the cloud coverage area, determining the best matching block of the remote sensing ground object, repairing the cloud coverage area by using the known matching block, rechecking the thin cloud coverage area, smoothing and filtering, and retaining the final result.
According to one aspect of the invention, an automatic processing method for cloud coverage area restoration of a single-time phase optical remote sensing image is provided, which is used for emergency application of cloud coverage area restoration and comprises the following steps:
firstly, in a thick cloud coverage area extraction stage, an extraction method based on color and texture features is provided, the thick cloud coverage area is preliminarily screened out by utilizing the color information of the thick cloud coverage area in an RGB color space, and then the extraction of the thick cloud coverage area is completed by utilizing the texture features reflected by the contrast of a symbiotic matrix; and then, in a cloud coverage area repairing stage, an improved Criminisi algorithm is provided to repair a thick cloud coverage area and a thin cloud coverage area, and the method can efficiently complete cloud coverage area repairing without multi-time phase data, historical data and multi-source data, so that rapid and reliable cloud coverage area repairing emergency application of the optical remote sensing image is realized.
Drawings
Fig. 1 is a flowchart of an automatic processing method for repairing a cloud coverage area of a single-phase optical remote sensing image according to an embodiment of the present invention.
Detailed Description
The following description is provided to illustrate how the methods provided by the present invention may be practiced.
Fig. 1 is a flowchart of an automatic processing method for repairing a cloud coverage area of a single-phase optical remote sensing image according to an embodiment of the present invention, where the method includes:
the first step is as follows: thick cloud coverage area extraction based on color and texture features
According to the distribution characteristics of a thick cloud coverage area in a large-view-field optical remote sensing image in an RGB color space, preliminary screening is carried out on a suspected thick cloud coverage area, and a candidate thick cloud coverage area is obtained. Secondly, extracting image texture characteristics based on a gray level co-occurrence matrix, screening out candidate areas with the texture characteristics of a thick cloud coverage area from color characteristics, and confirming the candidate areas, wherein the specific steps comprise:
step (1.1) of preliminary screening of thick cloud coverage areas based on color information
Because the highlight color characteristic of the cloud coverage area in the large-field-of-view optical remote sensing image is the most obvious characteristic attribute, the comprehensive judgment of the values of all components of the RGB color space can be used as the condition of primary screening. And traversing the image pixel by pixel, and taking the pixel value areas of the R, G, B three channels of the color image which are all larger than 210 as the coverage areas of the suspected thick clouds.
Step (1.2) identification of candidate cloud coverage areas based on texture features
Interference of some highlight ground objects may be included in the candidate thick cloud coverage area obtained from the whole remote sensing image in the step (1.1). And extracting image texture features based on the gray level co-occurrence matrix for the candidate regions in order to eliminate false alarm interference. In order to describe the texture characteristics of the candidate region quickly and effectively, a parameter which can reflect the texture characteristics of the cloud coverage area intuitively, namely the contrast ratio of the co-occurrence matrix, is mainly adopted in the derivative characteristics of the co-occurrence matrix. The method reflects the high-frequency detail degree and the texture groove depth degree of the image, and eliminates the non-cloud highlight area with over-high contrast of the co-occurrence matrix by utilizing the characteristics that the contrast of the co-occurrence matrix is smaller, the high-frequency detail is less and the groove is shallower according to the characteristic that the local texture of the cloud coverage area generally has smooth transition. And realizing candidate cloud coverage area confirmation.
The second step is that: cloud coverage area repair based on improved Criminisi algorithm
And based on the thick cloud coverage area screened in the last step, repairing the cloud coverage area by adopting an improved Criminisi algorithm. The algorithm mainly comprises five parts of determination of cloud coverage area removal priority, determination of remote sensing ground feature optimal matching block, cloud coverage area repair by utilizing known matching block, thin cloud coverage area recheck, smooth filtering and final result retention:
determination of cloud coverage removal priority in step (2.1)
For the suspected thick cloud coverage area obtained in the last step, the removal sequence needs to be determined, so that the linear structure propagation in the image is ensured, and the target boundaries are communicated. The priority calculation formula of the block with the target area edge point P as the center is as follows:
P(p)＝C(p)D(p)
wherein: c (p) is a confidence term, D (p) is a data term, and is defined as follows:
wherein: l Ψp| is the area of the currently selected block, α is the image normalization factor, npIs a unit normal vector at p on the edge of the target region, IpPoint p is an isolux line. In the initial stage c (p) ═ 0, the point with the highest priority is selected as the point to be removed.
Step (2.2) determination of best matching block of remote sensing ground feature
According to the principle of the least sum of squared difference distance, finding the optimal remote sensing ground feature texture matching block in the known area, wherein the formula is as follows:
Ψq＝mind(Ψp,Ψq)
wherein: ΨqIs found in the region with ΨpThe most suitable matching block.
Step (2.3) cloud coverage area repair with known matching blocks
Copying the remote sensing ground object optimal matching block obtained in the step (2.2) as an object to a corresponding target area position, updating the boundary and the confidence value of the target area, sequentially searching the target to be removed and the optimal matching block, and finally obtaining the image with the repaired thick cloud coverage area.
Coverage area rechecking of thin clouds in step (2.4)
According to the steps, the suspected thick cloud candidate area is screened out according to the color and texture characteristics in the steps (1.1) and (1.2), and the effective extraction of the thick cloud coverage area is realized. However, in the optical remote sensing image, besides the thick cloud coverage, there is also a thin cloud coverage. Due to the fact that image characterization characteristics of the remote sensing ground features are different, the identification of the remote sensing ground features can be influenced by the existence of some thin cloud coverage areas. Therefore, to meet the needs of emergency treatment, this patent uses a thin cloud coverage review strategy to overcome this problem. The method comprises the following specific steps:
2.4.1 first, the suspected thin cloud coverage area is determined. And (3) taking the image which does not contain the thick cloud after the repair of the thick cloud coverage area is finished in the step (2.3), and taking the area which is in the RGB color space and has R, G, B three channels which are all larger than 190 pixel values and accords with the texture characteristics of the cloud coverage area as a thin cloud coverage candidate area (the specific operation is the same as the steps 1,1 and 1, 2).
And 2.4.2, carrying out area-to-area statistics on the suspected thin cloud coverage area. And calculating the ratio of the thin cloud coverage area in the current image to the total image area as a comparison threshold T.
And 2.4.3, carrying out strategy division processing according to the area threshold value. If the value is less than the threshold value T (the value is 10 percent in the embodiment), the influence on the visual interpretation is considered to be limited, only local repair is needed, and the repair method is to directly carry out improved Criminisi algorithm repair (the specific operation is the same as the steps from 2.1 to 2.4); if the value is larger than the threshold value T, the influence on the visual interpretation cannot be objectively interpreted, the thin cloud repair is not carried out, and the 2.3-step thick cloud judgment result is directly output as a final result.
Step (2.5) smoothing and filtering and retaining the final result
Because some noise points caused by nonlinear restoration are always generated in the image repaired in the last step, visibility is improved by selecting a Gaussian smoothing filter for removal, and the image repaired is finally obtained.
Compared with the existing repairing method, the invention has the following advantages:
(1) aiming at the complex preconditions that the existing method mostly needs multi-temporal data in the same area, data of different sources in the same area, prior information of historical data and the like in the aspect of repairing the cloud coverage area of the optical remote sensing image, the patent provides an automatic processing method for repairing the cloud coverage area of the single-temporal optical remote sensing image. The method fully utilizes the color characteristics and the texture characteristics of the cloud coverage area to screen the cloud coverage candidate area, and then proposes a cloud coverage area repairing method based on an improved Criminisi algorithm to repair the cloud coverage area. The cloud coverage area restoration method can directly and effectively carry out cloud coverage area restoration only based on single-time-phase remote sensing image analysis and processing. The cloud coverage area restoration can be efficiently completed without multi-temporal data, historical data and multi-source data, and the rapid and reliable cloud coverage area restoration emergency application of the optical remote sensing image is realized.
(2) In the stage of screening the cloud coverage candidate area, the patent provides a suspected cloud coverage area screening method based on color and texture feature description. Firstly, an RGB space is utilized, and R, G, B three-channel pixel decision threshold values are set as conditions of primary screening. Then, a parameter of symbiotic matrix contrast which can visually reflect the texture characteristics of the cloud coverage area in the derivative characteristics of the gray level symbiotic matrix is adopted, the non-cloud coverage area highlight remote sensing ground objects are excluded, and the cloud coverage area to be repaired is further determined. The method can quickly and reliably select the suspected cloud coverage area from the large-field-of-view optical remote sensing image.
(3) In the cloud coverage candidate area repairing stage, the patent provides a cloud coverage candidate area repairing method based on an improved Criminisi algorithm. Firstly, the removal priority of the thick cloud coverage area is determined, and a proper removal sequence is set. And then, the optimal matching block of the remote sensing ground object is determined, so that the accuracy of the repairing effect is ensured. And then repairing the thick cloud coverage area by using the known matching block. And then, rechecking the thin cloud coverage area, repairing the coverage area with a small amount of thin clouds, and finally, performing smooth filtering and retaining the final result. The embodiment proves that the method can be used for rapidly repairing the suspected cloud coverage area, and the accuracy of repairing the cloud coverage area is improved.
Claims (1)
1. An automatic processing method for cloud coverage area restoration of a single-time phase optical remote sensing image is used for emergency application of cloud coverage area restoration and comprises the following steps:
firstly, in a thick cloud coverage area extraction stage, an extraction method based on color and texture features is provided, the thick cloud coverage area is preliminarily screened out by utilizing the color information of the thick cloud coverage area in an RGB color space, and then the extraction of the thick cloud coverage area is completed by utilizing the texture features reflected by the contrast of a symbiotic matrix; then, in a cloud coverage area repairing stage, an improved Criminisi algorithm is provided to repair a thick cloud coverage area and a thin cloud coverage area, and the method can efficiently complete cloud coverage area repairing without multi-temporal data, historical data and multi-source data, so that rapid and reliable cloud coverage area repairing emergency application of the optical remote sensing image is realized;
in the aspect of thick cloud coverage area extraction, firstly, an RGB color space is used for carrying out primary screening based on color characteristics on a suspected thick cloud coverage area, and then, a gray level co-occurrence matrix contrast ratio parameter is used for confirming a candidate cloud coverage area based on texture characteristics on the thick cloud coverage area, and the method specifically comprises the following steps:
A) based on the color information, preliminarily screening thick cloud coverage areas, including:
comprehensively judging the values of all components of the RGB color space of the whole remote sensing image as the condition of primary screening, traversing the image pixel by pixel, taking the pixel value areas of which the three channels of the component R, G, B are all more than 210 in the color image as the coverage area of the suspected thick cloud,
B) based on the texture features, confirming candidate cloud coverage areas, comprising:
based on the suspected thick cloud coverage area extracted in the step A), adopting a co-occurrence matrix contrast parameter in the derived characteristics of the co-occurrence matrix, and according to the characteristic that the local texture of the cloud coverage area generally has smooth transition, utilizing the characteristics that the co-occurrence matrix contrast is smaller, the high-frequency details are less and the furrow is shallower, excluding the non-cloud highlight area with the excessively high co-occurrence matrix contrast, and confirming the candidate cloud coverage area;
in a cloud coverage candidate area repairing stage, a cloud coverage candidate area repairing method based on an improved Criminisi algorithm is provided, and a step of judging the necessity of emergency repair of a thin cloud coverage area is added while the repair of a thick cloud coverage area is ensured, wherein the method specifically comprises the following steps:
1) determining a cloud coverage removal priority, comprising:
and determining a removal sequence aiming at the obtained suspected thick cloud coverage area to ensure that the linear structure in the image is propagated and the target boundary is communicated, wherein the priority calculation formula of the block taking the edge point P of the target area as the center is as follows:
P(p)＝C(p)D(p)
wherein: c (p) is a confidence term, D (p) is a data term, and is defined as follows:
wherein: l Ψp| is the area of the currently selected block, α is the image normalization factor, npIs a unit normal vector at p on the edge of the target region, IpThe initial stage C (p) is 0, the point with the highest priority is selected as the point to be removed,
2) determining a remote sensing ground object optimal matching block, comprising:
and finding the optimal remote sensing ground object texture matching block in the known region according to the principle of the least sum of squared differences, wherein the formula is as follows:
Ψq＝min d(Ψp,Ψq)
wherein: ΨqIs found in the region with ΨpThe most suitable matching block is the one that is,
3) repairing the cloud coverage area using the known matching blocks, comprising:
copying the optimal matching block of the remote sensing ground object obtained in the step 2) to the corresponding target area position, updating the boundary and the confidence value of the target area, sequentially searching the target to be removed and the optimal matching block to obtain an image with a repaired thick cloud coverage area,
4) performing thin cloud coverage recheck, comprising:
taking the image which is not contained with the thick cloud after the thick cloud coverage area is repaired in the step 3), taking R, G, B three channels in the RGB color space of the image which are all larger than 190 pixel values and according to the characteristic that the local texture of the cloud coverage area generally has smooth transition, and utilizing the characteristics that the contrast of the co-occurrence matrix is smaller, the high-frequency details are less and the furrow is shallower to eliminate the non-cloud highlight area with the over-high contrast of the co-occurrence matrix as the thin cloud coverage candidate area,
carrying out area ratio statistics on the suspected thin cloud coverage area, calculating the ratio of the thin cloud coverage area in the current image to the total image area as a comparison threshold T,
performing strategy division processing according to an area threshold, if the area threshold is smaller than a threshold T, considering that the influence on visual interpretation is limited, only local repair is needed, and the repair method is to directly perform improved Criminisi algorithm repair, and the specific operation is the same as the steps 1) to 4); if the value is larger than the threshold value T, the influence on the visual interpretation cannot be objectively interpreted, the thin cloud repair is not carried out, the thick cloud judgment result in the step 3) is directly output,
5) smoothing the filter and preserving the final result
Based on the image repaired in the step 4), noise points are removed by selecting a Gaussian smoothing filter, visibility is improved, and the image repaired is finally obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810018545.3A CN108280810B (en) | 2018-01-09 | 2018-01-09 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810018545.3A CN108280810B (en) | 2018-01-09 | 2018-01-09 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108280810A CN108280810A (en) | 2018-07-13 |
| CN108280810B true CN108280810B (en) | 2020-08-14 |
Family
ID=62803328
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810018545.3A Active CN108280810B (en) | 2018-01-09 | 2018-01-09 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108280810B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111259B (en) * | 2019-05-15 | 2021-03-02 | 电子科技大学 | Multi-source image fusion method based on region guidance |
| CN111797744B (en) * | 2020-06-28 | 2022-06-10 | 武汉大学 | Multimode remote sensing image matching method based on co-occurrence filtering algorithm |
| CN112561830B (en) * | 2020-12-23 | 2022-11-18 | 安徽大学 | Endoscope image highlight repair method and device |
| CN113298808B (en) * | 2021-06-22 | 2022-03-18 | 哈尔滨工程大学 | Method for repairing building shielding information in tilt-oriented remote sensing image |
| CN116245757B (en) * | 2023-02-08 | 2023-09-19 | 北京艾尔思时代科技有限公司 | Multi-scene universal remote sensing image cloud restoration method and system for multi-mode data |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902956A (en) * | 2012-09-10 | 2013-01-30 | 中国人民解放军理工大学气象学院 | Ground-based visible cloud image recognition processing method |
| CN104484670A (en) * | 2014-10-24 | 2015-04-01 | 西安电子科技大学 | Remote sensing image cloud detection method based on pseudo color and support vector machine |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080031702A (en) * | 2008-02-22 | 2008-04-10 | 강용균 | Method of cloud masking in remote sensing images |
| CN103020939B (en) * | 2012-12-18 | 2015-06-10 | 武汉大学 | Method for removing large-area thick clouds for optical remote sensing images through multi-temporal data |
| CN103208001B (en) * | 2013-02-06 | 2016-12-28 | 华南师范大学 | In conjunction with shape-adaptive neighborhood and the remote sensing image processing method of texture feature extraction |
| CN103472501B (en) * | 2013-09-06 | 2016-05-25 | 中国气象科学研究院 | Cloud detection and all-sky total amount of cloud detection method and system |
| CN104778668B (en) * | 2015-04-16 | 2017-12-15 | 电子科技大学 | The thin cloud minimizing technology of remote sensing image based on visible light wave range spectrum statistical nature |
| CN105426903A (en) * | 2015-10-27 | 2016-03-23 | 航天恒星科技有限公司 | Cloud determination method and system for remote sensing satellite images |
| CN105354865B (en) * | 2015-10-27 | 2018-01-26 | 武汉大学 | The automatic cloud detection method of optic of multispectral remote sensing satellite image and system |
-
2018
- 2018-01-09 CN CN201810018545.3A patent/CN108280810B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902956A (en) * | 2012-09-10 | 2013-01-30 | 中国人民解放军理工大学气象学院 | Ground-based visible cloud image recognition processing method |
| CN104484670A (en) * | 2014-10-24 | 2015-04-01 | 西安电子科技大学 | Remote sensing image cloud detection method based on pseudo color and support vector machine |
Non-Patent Citations (1)
| Title |
|---|
| Retrieval of Remote Sensing Images Using Color,Texture and Spectral Features;P.M et al;《International Journal of Engineering Science and Technology》;20100930;第2卷(第9期);第4306-4311页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108280810A (en) | 2018-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108280810B (en) | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image | |
| CN108876856B (en) | Fire source identification and positioning method and system for large building | |
| CN111415363B (en) | Image edge identification method | |
| CN106296670B (en) | A kind of Edge detection of infrared image based on the watershed Retinex--Canny operator | |
| CN111339948B (en) | Automatic identification method for newly-added buildings of high-resolution remote sensing images | |
| CN104318546A (en) | Multi-scale analysis-based greenhouse field plant leaf margin extraction method and system | |
| CN103839267A (en) | Building extracting method based on morphological building indexes | |
| Liu et al. | Robust image-based crack detection in concrete structure using multi-scale enhancement and visual features | |
| CN111310771B (en) | Road image extraction method, device and equipment of remote sensing image and storage medium | |
| CN114764801A (en) | Weak and small ship target fusion detection method and device based on multi-vision significant features | |
| CN110348442B (en) | Shipborne radar image offshore oil film identification method based on support vector machine | |
| CN107038690B (en) | Moving shadow removing method based on multi-feature fusion | |
| Ren et al. | Effective SAR sea ice image segmentation and touch floe separation using a combined multi-stage approach | |
| Sidike et al. | Automatic building change detection through adaptive local textural features and sequential background removal | |
| Wu et al. | Research on crack detection algorithm of asphalt pavement | |
| CN110298816B (en) | Bridge crack detection method based on image regeneration | |
| Cheng et al. | Power pole detection based on graph cut | |
| Ji et al. | A novel method of image segmentation using watershed transformation | |
| Khan et al. | Shadow removal from digital images using multi-channel binarization and shadow matting | |
| CN112686222B (en) | Method and system for detecting ship target by satellite-borne visible light detector | |
| CN107609537A (en) | A kind of waterfront line detecting method based on HSV space Surface Picture feature | |
| Herumurti et al. | Automatic urban road extraction on DSM data based on fuzzy ART, region growing, morphological operations and Radon transform | |
| Ye et al. | Research on Automatic Highway Extraction Technology Based on Spectral Information of Remote Sensing Images. | |
| Wang et al. | A model for target acquisition and edge detection under complex scenes | |
| CN114022473B (en) | Horizon detection method based on infrared image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |