CN109886904B - SAR image and low-resolution multispectral image fusion method and system - Google Patents
SAR image and low-resolution multispectral image fusion method and system Download PDFInfo
- Publication number
- CN109886904B CN109886904B CN201910071324.7A CN201910071324A CN109886904B CN 109886904 B CN109886904 B CN 109886904B CN 201910071324 A CN201910071324 A CN 201910071324A CN 109886904 B CN109886904 B CN 109886904B
- Authority
- CN
- China
- Prior art keywords
- image
- resolution
- sar
- mean
- multispectral
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for fusing an SAR image and a low-resolution multispectral image, wherein the method comprises the following steps: calculating a spatial resolution ratio k of the low-resolution multispectral image and the high-resolution SAR image; based on the spatial resolution ratio k, carrying out bilinear interpolation on the low-resolution multispectral image to obtain an up-sampling multispectral image; based on the spatial resolution ratio k, carrying out mean value filtering on the high-resolution SAR image to obtain a spatial detail image of the SAR image; and transforming the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image. The invention can effectively keep the backscattering characteristic of the high-resolution SAR image and the hue and saturation of the multispectral image, improves the visual interpretation of the SAR image, and can be used for automatically producing the fusion image.
Description
Technical Field
The invention relates to a remote sensing image fusion method, in particular to a visual interpretation-oriented high-resolution SAR image and low-resolution multispectral image fusion method, and belongs to the technical field of digital image processing.
Background
Synthetic Aperture Radar (SAR) is a microwave active imaging Radar capable of realizing high resolution, has all-time and all-weather inherent advantages, plays an important role in disaster detection, environmental monitoring, military reconnaissance and other aspects, and is generally regarded by various countries. However, the SAR image is very different from the optical visual image seen in daily life of people, so that the intuitive interpretability is poor, and the SAR image interpretation is very difficult. The current SAR image interpretation mainly depends on experienced interpretation experts, but with the further improvement of image resolution, the SAR imaging capability gradually advances from 'general survey' to 'detailed survey', and an interpretation application mode mainly based on expert interpretation is difficult to meet the application requirements of various industries on SAR images.
To overcome the above-mentioned drawbacks, research on remote sensing image fusion technology has been developed. At present, remote sensing image fusion technology mainly focuses on fusion of panchromatic and multispectral and panchromatic and hyperspectral images, and the methods can be divided into two categories of additive transformation and multiplicative transformation. The additive transformation fusion mainly comprises fusion methods based on IHS transformation, PCA transformation, GS transformation, wavelet transformation and the like. The multiplicative transformation fusion method mainly comprises a Brovey transformation method, a UNB-Panship fusion method and the like.
On one hand, however, the imaging difference between the SAR image and the optical remote sensing image is very large, so that the fusion method for panchromatic and multispectral images and panchromatic and hyperspectral images is difficult to meet the fusion requirement of the SAR image and multispectral images, and the fusion effect is poor. On the other hand, the existing fusion method of the SAR image and the multispectral image mainly utilizes high-resolution space detail information of the multispectral image to sharpen the SAR image, so that a target in the SAR image is clearer. However, the multispectral image mainly records the spectral information of the ground object, and the SAR image records the backscattering electromagnetic wave capability of the ground object, and the spatial detail information of the multispectral image and the SAR image has a large difference, so that sharpening the SAR image by the multispectral image often causes blurring of the fused image due to inconsistent spatial detail.
Disclosure of Invention
In order to solve the problems, the invention provides a method and a system for fusing a high-resolution SAR image and a low-resolution multispectral image, which are oriented to visual interpretation, and spectral color information of the low-resolution multispectral image is embedded into the high-resolution SAR image, so that the visual interpretation of the SAR image is improved.
In order to achieve the purpose, the technical scheme provided by the invention is as follows:
a method for fusing an SAR image and a low-resolution multispectral image is characterized by comprising the following steps:
calculating a spatial resolution ratio k of the low-resolution multispectral image and the high-resolution SAR image;
based on the spatial resolution ratio k, carrying out bilinear interpolation on the low-resolution multispectral image to obtain an up-sampling multispectral image;
based on the spatial resolution ratio k, carrying out mean value filtering on the high-resolution SAR image to obtain a spatial detail image of the SAR image;
and transforming the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image.
Further, the bilinear interpolation includes linearly interpolating the low resolution multispectral image by rows at a k-fold ratio and linearly interpolating by columns at a k-fold ratio.
Further, a ratio of the high-resolution SAR image to the SAR average filtering image is calculated to obtain a space detail image.
And further, carrying out mean filtering on the high-resolution SAR image by taking 1/k as a smoothing factor to generate an SAR mean filtering image.
Further, the high-resolution SAR and low-resolution multispectral fusion image is generated by multiplicative transformation of the up-sampling multispectral image and the spatial detail image.
The invention also provides a system for fusing the high-resolution SAR image and the low-resolution multispectral image, which comprises the following steps:
the first input module is used for inputting the low-resolution multispectral image;
the second input module is used for inputting a high-resolution SAR image;
the spatial resolution ratio calculation module is used for calculating the spatial resolution ratio of the low-resolution multispectral image and the high-resolution SAR image;
the up-sampling multispectral image acquisition module is used for performing bilinear interpolation on the low-resolution multispectral image based on the spatial resolution ratio to obtain an up-sampling multispectral image;
the SAR space detail image acquisition module is used for carrying out mean value filtering on the high-resolution SAR image based on the space resolution ratio to acquire a space detail image of the SAR image;
and the image fusion module is used for performing multiplicative transformation on the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image.
Further, the system further comprises an output module for outputting the high-resolution SAR and low-resolution multispectral fusion image.
Further, the output module comprises a sending module and/or a display module.
Due to the adoption of the scheme, the invention has the following technical effects:
(1) the generated fusion image can effectively keep the backscattering characteristic of the high-resolution SAR image and the hue and saturation of the multispectral image, and the intuitionistic interpretation of the SAR image is improved.
(2) The method has good adaptability to high-resolution SAR images and low-resolution multispectral images acquired by different satellites, does not need to manually set fusion parameters, and can be used for automatically producing fusion images.
Drawings
FIG. 1 is a flow chart of a method for fusing a high resolution SAR image with a low resolution multispectral image according to the present invention;
fig. 2(a) -2 (c) show fused images generated by the image fusion method employed. Fig. 2(a) is a low-resolution multispectral image, fig. 2(b) is a high-resolution SAR image, and fig. 2(c) is a generated fusion image.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
As shown in fig. 1, the registered high-resolution SAR image and the low-resolution multispectral image are fused sequentially through the following steps:
step 1: and calculating the spatial resolution ratio of the low-resolution multispectral image to the high-resolution SAR image.
Suppose the spatial resolution of the high-resolution SAR image is d1Rice, low resolutionThe spatial resolution of the rate-multispectral image is d2And if the ratio k of the spatial resolution of the low-resolution multispectral image to the spatial resolution of the high-resolution SAR image is meter, the ratio k is as follows:
k＝d2/d1
step 2: and performing bilinear interpolation on the low-resolution multispectral image based on the spatial resolution ratio to obtain an up-sampling multispectral image.
The main purpose of upsampling (upsampling) an image is to enlarge the original image so that it can be displayed with a higher resolution, and the basic method is to insert new elements between pixels by using a suitable interpolation algorithm on the basis of the pixels of the original image.
The existing image bilinear interpolation usually adopts a four-point method, the calculation complexity is high, and the size of a standard remote sensing image is usually larger than 10000 multiplied by 10000 pixels, so that the whole image interpolation calculation is time-consuming.
The invention decomposes bilinear interpolation into two linear interpolations, namely: firstly, carrying out linear interpolation on a low-resolution multispectral image according to lines at a k-fold ratio; linear interpolation is then performed column by column at the k-fold ratio. Those skilled in the art will readily understand that the above-mentioned two linear interpolations can also be performed first on the low-resolution multispectral image by linear interpolation in columns at k-fold ratio; linear interpolation is then performed in rows at a k-fold ratio.
For example, assuming a low resolution multispectral image with T pixels, the upsampled multispectral image has k2T pixels. If conventional bilinear interpolation is used, there is 8T (k)2-1) floating-point number multiplications and 3T (k)2-1) floating point number addition. With the interpolation method of the present invention, only (k) is used2-1) T floating-point number multiplications and (k)2-1) T floating-point number additions, thus reducing the computational effort considerably.
And step 3: and performing mean filtering on the high-resolution SAR image based on the spatial resolution ratio to obtain a spatial detail image of the SAR image.
The method comprises the following substeps:
step 31: and carrying out average filtering on the high-resolution SAR image to generate an SAR average filtering image.
The method takes 1/k as a smoothing factor to carry out mean filtering on the high-resolution SAR image so as to generate the SAR mean filtering image.
Specifically, for the high-resolution SAR image, mean filtering is performed by rows by taking 1/k as a smoothing factor, and then mean filtering is performed by columns by taking 1/k as a smoothing factor, so that the SAR mean filtering image is generated. One skilled in the art will readily appreciate that the SAR mean filtered image may also be generated by first performing a column-wise mean filtering with 1/k as a smoothing factor and then performing a row-wise mean filtering with 1/k as a smoothing factor.
Taking "first mean filtering by rows with 1/k as a smoothing factor, and then mean filtering by columns with 1/k as a smoothing factor" as an example, assume that S (i, j) is the pixel value of the ith row and jth column of the high-resolution SAR image, and Sr(i, j) is the pixel value of S (i, j) after line mean filtering,
is Sr(i, j) column-mean filtered pixel values, then:
(1) when k is an integer, filtering S (i, j) by line mean to obtain:
to Sr(i, j) filtering by column mean to obtain:
generated image
Namely, the average filtering image of the high-resolution SAR image.
(2) When k is not an integer, filtering S (i, j) according to a line mean value, firstly calculating a pixel value S (i, j + k-1) of the ith line and the j + k-1 column of the high-resolution SAR image by using linear interpolation, wherein the calculation formula is as follows:
then, filtering S (i, j) by line mean to obtain:
to Sr(i, j) when filtering by column mean, first calculate the image S by linear interpolationrPixel value S of j column of i + k-1 rowr(i + k-1, j), which is calculated as follows:
then, filtering by columns is performed to obtain:
generated image
Namely, the average filtering image of the high-resolution SAR image.
Step 32: and calculating the ratio of the high-resolution SAR image to the SAR average filtering image to obtain a space detail image.
Let S (i, j) be the pixel value of the ith row and the jth column of the high-resolution SAR image,
calculating the spatial detail image by adopting the following modes, wherein R (i, j) is the pixel value of the ith row and the jth column of the mean filtering image of the high-resolution SAR image, and R (i, j) is the pixel value of the ith row and the jth column of the spatial detail image:
although the embodiment of the present invention describes "step 2" and "step 3" for convenience of explanation, this does not mean that the present invention defines the order of step 2 and step 3. It is understood by those skilled in the art that the sequence of step 2 and step 3 may be changed, or according to another embodiment, step 2 and step 3 may be performed simultaneously.
And 4, step 4: and performing multiplicative transformation on the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image.
Suppose Mu(i, j) is the pixel value of the ith row and the jth column of the u wave band of the low resolution multispectral image, R (i, j) is the pixel value of the ith row and the jth column of the spatial detail image, Fu(i, j) is the pixel value of the ith row and the jth column of the u wave band of the fused image, and the fused image is generated by adopting multiplicative transformation as follows:
Fu(i,j)＝R(i,j)×Mu(i,j)
by adopting the method provided by the invention, the multispectral image with the resolution of 2.4 meters and the SAR image with the resolution of 0.5 meter are fused, and the fusion result is shown in figure 2. Fig. 2(a) is a low-resolution multispectral image, fig. 2(b) is a high-resolution SAR image, and fig. 2(c) is a generated fusion image. The observation of the fused image is visible, and the visual interpretation degree of the SAR image is greatly improved by fusing the spectral colors into the SAR image.
To further analyze the spectral color fidelity of the fused image, the fused image and the multispectral image are transformed into IHS color space. Table 1 lists the IHS color space contrast of the fused image and the multispectral image generated using the method of the present invention. Wherein M isuAnd (i, j) is the pixel value of the ith row and the jth column of the u wave band of the low-resolution multispectral image, and R (i, j) is the pixel value of the ith row and the jth column of the spatial detail image.
As can be seen from table 1, the fused image completely agrees with the full-color image in the H component and the S component of the color of the drawn image. The method disclosed by the invention has the advantages that when the spectral color information of the low-resolution multispectral image is fused into the high-resolution SAR image, the spectral color of the multispectral image is not changed, and people can understand the high-resolution SAR image by using the spectral color.
TABLE 1 fused image vs. multispectral image in IHS color space contrast
According to the visual interpretation-oriented high-resolution SAR and low-resolution multispectral image fusion method, the spectral color information of the low-resolution multispectral image is fused into the high-resolution SAR image, so that the visual interpretation of the SAR image is improved. However, obviously, the fusion method described in this specification is also applicable to fusion operation of other electromagnetic wave imaging devices and optical imaging devices to acquire images, and the obtained beneficial effects are also similar.
The invention also provides a system for fusing the high-resolution SAR image and the low-resolution multispectral image, which comprises the following steps:
the first input module is used for inputting the low-resolution multispectral image;
the second input module is used for inputting a high-resolution SAR image;
the spatial resolution ratio calculation module is used for calculating the spatial resolution ratio of the low-resolution multispectral image and the high-resolution SAR image;
the up-sampling multispectral image acquisition module is used for performing bilinear interpolation on the low-resolution multispectral image based on the spatial resolution ratio to obtain an up-sampling multispectral image;
the SAR space detail image acquisition module is used for carrying out mean value filtering on the high-resolution SAR image based on the space resolution ratio to acquire a space detail image of the SAR image;
and the image fusion module is used for performing multiplicative transformation on the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image.
Further, the high-resolution SAR image and low-resolution multispectral image fusion system further comprises an output module for outputting the high-resolution SAR and low-resolution multispectral fusion image. The output module comprises a sending module and/or a display module.
Although the embodiments of the present invention have been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the embodiments of the present invention.
Claims (6)
1. A method for fusing an SAR image and a low-resolution multispectral image is characterized by comprising the following steps:
calculating a spatial resolution ratio k of the low-resolution multispectral image and the high-resolution SAR image;
based on the spatial resolution ratio k, carrying out bilinear interpolation on the low-resolution multispectral image to obtain an up-sampling multispectral image;
taking 1/k as a smoothing factor, carrying out mean filtering on the high-resolution SAR image to generate an SAR mean filtering image, and specifically comprising the following steps:
expressing the pixel value of the ith row and the jth column of the high-resolution SAR image by S (i, j), Sr(i, j) represents the pixel value of S (i, j) after line mean filtering,
denotes Sr(i, j) column-mean filtered pixel values;
(1) when k is an integer, filtering S (i, j) by line mean to obtain:
to Sr(i, j) filtering by column mean to obtain:
then the resulting image is filtered by column mean
A mean filtered image which is a high resolution SAR image;
(2) when k is not an integer, filtering S (i, j) according to a line mean value, firstly calculating a pixel value S (i, j + k-1) of the ith line and the j + k-1 column of the high-resolution SAR image by using linear interpolation, wherein the calculation formula is as follows:
then, filtering S (i, j) by line mean to obtain:
to Sr(i, j) when filtering by column mean, first calculate the image S by linear interpolationrPixel value S of j column of i + k-1 rowr(i + k-1, j), which is calculated as follows:
then, performing a column-wise mean filtering to obtain:
then the resulting image is filtered by column mean
A mean filtered image which is a high resolution SAR image;
calculating the ratio of the high-resolution SAR image to the SAR average filtering image to obtain a space detail image of the SAR image;
and transforming the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image.
2. The method for fusing the SAR image and the low-resolution multispectral image according to claim 1, wherein the method comprises the following steps: the bilinear interpolation includes linear interpolation of low resolution multispectral images by rows at a k-fold ratio and linear interpolation by columns at a k-fold ratio.
3. The method for fusing the SAR image and the low-resolution multispectral image according to claim 1, wherein the method comprises the following steps: and performing multiplicative transformation on the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image.
4. A high-resolution SAR image and low-resolution multispectral image fusion system is characterized by comprising:
the first input module is used for inputting the low-resolution multispectral image;
the second input module is used for inputting a high-resolution SAR image;
the spatial resolution ratio calculation module is used for calculating the spatial resolution ratio k of the low-resolution multispectral image and the high-resolution SAR image;
the up-sampling multispectral image acquisition module is used for performing bilinear interpolation on the low-resolution multispectral image based on the spatial resolution ratio k to obtain an up-sampling multispectral image;
the SAR space detail image acquisition module performs mean filtering on the high-resolution SAR image by taking 1/k as a smoothing factor to generate an SAR mean filtering image, and specifically comprises the following steps:
expressing the pixel value of the ith row and the jth column of the high-resolution SAR image by S (i, j), Sr(i, j) represents the pixel value of S (i, j) after line mean filtering,
denotes Sr(i, j) column-mean filtered pixel values;
(1) when k is an integer, filtering S (i, j) by line mean to obtain:
to Sr(i, j) filtering by column mean to obtain:
then the resulting image is filtered by column mean
A mean filtered image which is a high resolution SAR image;
(2) when k is not an integer, filtering S (i, j) according to a line mean value, firstly calculating a pixel value S (i, j + k-1) of the ith line and the j + k-1 column of the high-resolution SAR image by using linear interpolation, wherein the calculation formula is as follows:
then, filtering S (i, j) by line mean to obtain:
to Sr(i, j) when filtering by column mean, first calculate the image S by linear interpolationrPixel value S of j column of i + k-1 rowr(i + k-1, j), which is calculated as follows:
then, performing a column-wise mean filtering to obtain:
then the resulting image is filtered by column mean
A mean filtered image which is a high resolution SAR image;
calculating the ratio of the high-resolution SAR image to the SAR average filtering image to obtain a space detail image of the SAR image;
and the image fusion module is used for performing multiplicative transformation on the up-sampling multispectral image and the spatial detail image to generate a high-resolution SAR and low-resolution multispectral fusion image.
5. The system according to claim 4, further comprising an output module for outputting the fused image of SAR and multispectral image with high resolution and low resolution.
6. The system for fusing the high-resolution SAR image and the low-resolution multispectral image according to claim 5, wherein the output module comprises a transmitting module and/or a display module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910071324.7A CN109886904B (en) | 2019-01-25 | 2019-01-25 | SAR image and low-resolution multispectral image fusion method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910071324.7A CN109886904B (en) | 2019-01-25 | 2019-01-25 | SAR image and low-resolution multispectral image fusion method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109886904A CN109886904A (en) | 2019-06-14 |
| CN109886904B true CN109886904B (en) | 2021-08-10 |
Family
ID=66926804
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910071324.7A Expired - Fee Related CN109886904B (en) | 2019-01-25 | 2019-01-25 | SAR image and low-resolution multispectral image fusion method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109886904B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113112533B (en) * | 2021-04-15 | 2022-05-03 | 宁波甬矩空间信息技术有限公司 | SAR-multispectral-hyperspectral integrated fusion method based on multiresolution analysis |
| CN117197018A (en) * | 2023-09-27 | 2023-12-08 | 北京航空航天大学 | Multiplication transformation remote sensing image fusion method for maintaining spectral distribution |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254311A (en) * | 2011-06-10 | 2011-11-23 | 中国科学院深圳先进技术研究院 | Method and system for fusing remote sensing images |
| CN103236047A (en) * | 2013-03-28 | 2013-08-07 | 北京航空航天大学 | Method for fusing full-color and multi-spectral images on basis of fitting for substituted components |
| CN104021536A (en) * | 2014-06-16 | 2014-09-03 | 西北工业大学 | Self-adaptation SAR image and multispectral image fusion method |
| CN107016641A (en) * | 2016-09-23 | 2017-08-04 | 北京航空航天大学 | A kind of panchromatic and hyperspectral image fusion method based on improvement ratio transformation |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102158653B (en) * | 2011-05-03 | 2013-01-16 | 东华大学 | Device and method for acquiring digital image with high dynamic range in real time |
| CN104700379B (en) * | 2014-12-29 | 2019-05-10 | 烟台大学 | A kind of remote sensing image fusion method based on Multiscale Morphological constituent analysis |
| CN105528772B (en) * | 2015-12-05 | 2018-10-16 | 中国航空工业集团公司洛阳电光设备研究所 | A kind of image interfusion method based on directiveness filtering |
| CN106251368B (en) * | 2016-06-12 | 2019-05-14 | 中国科学院遥感与数字地球研究所 | The fusion method of SAR image and multispectral image based on BEMD |
| US20180129368A1 (en) * | 2016-11-07 | 2018-05-10 | Microsoft Technology Licensing, Llc | Data characteristic guided filter builder |
| CN108549902B (en) * | 2018-03-14 | 2022-08-16 | 中国科学院遥感与数字地球研究所 | Improved SAR image and multi-spectrum optical image fusion method |
| CN109002851B (en) * | 2018-07-06 | 2021-04-02 | 东北大学 | Fruit classification method based on image multi-feature fusion and application |
| CN109146819A (en) * | 2018-08-29 | 2019-01-04 | 广州清宇信息科技有限公司 | A kind of denoising of SAR image and multi-sources RS data fusion algorithm |
-
2019
- 2019-01-25 CN CN201910071324.7A patent/CN109886904B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254311A (en) * | 2011-06-10 | 2011-11-23 | 中国科学院深圳先进技术研究院 | Method and system for fusing remote sensing images |
| CN103236047A (en) * | 2013-03-28 | 2013-08-07 | 北京航空航天大学 | Method for fusing full-color and multi-spectral images on basis of fitting for substituted components |
| CN104021536A (en) * | 2014-06-16 | 2014-09-03 | 西北工业大学 | Self-adaptation SAR image and multispectral image fusion method |
| CN107016641A (en) * | 2016-09-23 | 2017-08-04 | 北京航空航天大学 | A kind of panchromatic and hyperspectral image fusion method based on improvement ratio transformation |
Non-Patent Citations (2)
| Title |
|---|
| Hyperspectral and panchromatic image fusion through an improved ratio enhancement;Qizhi Xu 等;《JOURNAL OF APPLIED REMOTE SENSING》;20170308;第11卷;全文 * |
| 基于比值变换的全色与多光谱图像高保真融合方法;徐其志 等;《计算机科学》;20141031;第41卷(第10期);第19-22页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109886904A (en) | 2019-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Loncan et al. | Hyperspectral pansharpening: A review | |
| Luo et al. | Pansharpening via unsupervised convolutional neural networks | |
| Song et al. | Spatiotemporal satellite image fusion through one-pair image learning | |
| Wang et al. | A comparative analysis of image fusion methods | |
| González-Audícana et al. | A low computational-cost method to fuse IKONOS images using the spectral response function of its sensors | |
| Chen et al. | Image fusion with local spectral consistency and dynamic gradient sparsity | |
| Choi | A new intensity-hue-saturation fusion approach to image fusion with a tradeoff parameter | |
| CN106920214B (en) | Super-resolution reconstruction method for space target image | |
| Tuna et al. | Single-frame super resolution of remote-sensing images by convolutional neural networks | |
| CN110415199B (en) | Multispectral remote sensing image fusion method and device based on residual learning | |
| CN112419150B (en) | Image super-resolution reconstruction method of arbitrary multiple based on bilateral upsampling network | |
| Li et al. | Underwater image high definition display using the multilayer perceptron and color feature-based SRCNN | |
| CN111008936B (en) | Multispectral image panchromatic sharpening method | |
| CN109886904B (en) | SAR image and low-resolution multispectral image fusion method and system | |
| US20130051703A1 (en) | Method for detecting directions of regularity in a two-dimensional image | |
| Lakshman et al. | Image interpolation using shearlet based iterative refinement | |
| CN116309070A (en) | Super-resolution reconstruction method and device for hyperspectral remote sensing image and computer equipment | |
| CN111383203B (en) | Panchromatic and multispectral remote sensing image fusion method based on regional fitting | |
| Zhong et al. | Attention_FPNet: Two-branch remote sensing image pansharpening network based on attention feature fusion | |
| CN106846250A (en) | A kind of super resolution ratio reconstruction method based on multi-scale filtering | |
| Yocky | Artifacts in wavelet image merging | |
| CN112508832A (en) | Object-oriented remote sensing image data space-time fusion method, system and equipment | |
| Tiwari et al. | An Improved IHS Image Fusion Algorithm using Medoid Intensity Match and Bilateral Filter | |
| CN111523451A (en) | Method for constructing high-space-time resolution NDVI data | |
| Hemalatha et al. | Comparison of DWT, DWT-SWT, and DT-CWT for low resolution satellite images enhancement |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210810 Termination date: 20220125 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |