CN109816616B - Non-refrigeration infrared image and solar blind ultraviolet image fusion method based on tower decomposition - Google Patents
Non-refrigeration infrared image and solar blind ultraviolet image fusion method based on tower decomposition Download PDFInfo
- Publication number
- CN109816616B CN109816616B CN201811586175.XA CN201811586175A CN109816616B CN 109816616 B CN109816616 B CN 109816616B CN 201811586175 A CN201811586175 A CN 201811586175A CN 109816616 B CN109816616 B CN 109816616B
- Authority
- CN
- China
- Prior art keywords
- image
- layer
- pyramid
- fusion
- infrared image
- 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
Images
Abstract
The invention discloses a method for fusing an uncooled infrared image and a solar blind ultraviolet image based on tower decomposition. The method comprises the following steps: firstly, median filtering is carried out on an uncooled infrared image to remove salt and pepper noise, and contrast adjustment is realized through upper platform histogram equalization; then, respectively carrying out Gaussian pyramid decomposition and Laplacian pyramid decomposition on the uncooled infrared image and the uncooled ultraviolet image; then, fusing the third layer of the Laplacian pyramid by adopting weighted average, and fusing the second layer and the first layer by adopting gray value enlargement; and finally, upsampling the layered and fused Laplacian pyramid from the third layer, adding the second layer after Gaussian kernel convolution filtering, continuously upsampling the obtained image, adding the first layer after Gaussian kernel convolution filtering, finishing the reconstruction process, and obtaining the final fused image. The invention has the advantages of good image fusion effect, simple operation and less required resources.
Description
Technical Field
The invention belongs to the field of image processing, and particularly relates to a method for fusing an uncooled infrared image and a solar blind ultraviolet image based on tower decomposition.
Background
In an electric power system, the load borne by a transmission line continuously rises along with the enlargement of the national power grid scale. Because of the reduction of the insulating property of the transmission line, the transmission line exposed in the air can generate corona discharge, heat can be accumulated along with the increase of time, and if the corona discharge is not discovered in time and repaired, the property of people can be greatly lost, so that the high-performance corona detection device has wide application prospect and market demand.
In the process of corona discharge, after air is ionized by a strong electric field, a gas luminescence phenomenon can be generated in an ionization region. The ultraviolet spectrum section of the corona discharge is mainly concentrated in an ultraviolet region of 300-400 nm and a solar blind region of 230-280 nm. Because ultraviolet radiation of less than 280nm in sunlight is absorbed by the ozone layer when passing through the atmosphere, the wavelength range below 280nm is called a "solar dead zone". Ultraviolet radiation in a solar blind area detected near the ground surface is not interfered by sunlight, so that the method has high detectivity and sensitivity, but the ultraviolet image has poor distinguishability, and images of other wave bands are required to provide background references.
Disclosure of Invention
The invention aims to provide a method for fusing an uncooled infrared image and a solar blind ultraviolet image based on tower decomposition, which has the advantages of good fusion effect, simple operation and less required resources.
The technical solution for realizing the purpose of the invention is as follows: a method for fusing an uncooled infrared image and a solar blind ultraviolet image based on tower decomposition comprises the following steps:
acquiring an uncooled infrared image, firstly performing median filtering to remove salt and pepper noise, and then performing upper platform histogram equalization to improve the contrast of the whole image; respectively carrying out Gaussian pyramid decomposition and Laplacian pyramid decomposition on the uncooled infrared image and the ultraviolet image; in the fusion, the Laplace pyramid third layer adopts weighted average fusion, and the second layer and the first layer adopt gray value taking large fusion; and after hierarchical fusion, reconstructing the three layers of Laplacian pyramids to obtain a final fusion image.
Further, the method for fusing the uncooled infrared image and the solar blind ultraviolet image based on the tower decomposition comprises the following steps of:
step 1, performing median filtering on an uncooled infrared image to remove salt and pepper noise;
step 2, performing upper platform histogram equalization on the non-refrigeration infrared image subjected to median filtering;
step 3, taking the preprocessed uncooled infrared image as the first layer GT of the Gaussian pyramid 1 Using a Gaussian kernel g 3×3 To GT 1 Performing convolution and downsampling to remove even rows and columns to obtain a second layer GT of Gaussian pyramid 2 (ii) a By using a Gaussian kernel g 3×3 To GT 2 Convolution and down-sampling are carried out to obtain a third layer GT of a Gaussian pyramid 3 ；
Step 4, for GT in step 3 2 Upsampling, even rows and columns filled with 0, and then using Gaussian kernel g 3×3 Performing convolution to obtain
To GT 3 Upsampling, filling even rows and columns with 0, and then applying a Gaussian kernel g 3×3 Performing convolution to obtain
Use of
Obtaining a Laplacian pyramid first layer LT 1 Use of
Obtaining a Laplacian pyramid second layer LT 2 Laplacian pyramid third layer LT 3 And GT 3 The same;
and 5, repeating the step 3 and the step 4 for the solar blind ultraviolet image to obtain the Gaussian pyramid GU of the solar blind ultraviolet image i And Laplacian pyramid LU i ；
Step 6, for LT 3 And LU 3 Carrying out weighted average fusion by adopting the ways of uncooled infrared image weight 1/4 and solar blind ultraviolet image weight 3/4 to obtain a fused image LF 3 (ii) a For LT 2 And LU 2 、LT 1 And LU 1 Fusing in a way of increasing the gray value to obtain a fused image LF 2 And LF 1 ；
Step 7, fusing images LF 3 Upsampling and using Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 2 Adding, up-sampling the resulting images, and applying a Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 1 And adding, and finishing reconstruction to obtain a final fusion image.
Further, the Gaussian core g described in step 3 3×3 The method comprises the following steps:
further, the step 4 is described for GT in step 3 2 Upsampling, even rows and columns filled with 0, and then using Gaussian kernel g 3×3 Performing convolution to obtain
To GT 3 Upsampling, filling even rows and columns with 0, and then applying a Gaussian kernel g 3×3 Performing convolution to obtain
Use of
Obtaining a Laplacian pyramid first layer LT 1 Use of
Obtaining a Laplacian pyramid second layer LT 2 Laplacian pyramid third layer LT 3 And GT 3 The same is true, specifically as follows:
the process of constructing the laplacian pyramid is shown in formula (2):
where UP represents an UP-sampling operation and,
representing a convolution operation, g 3×3 Represents a gaussian kernel; third layer L of Laplacian pyramid 3 I.e. the third layer G of the Gaussian pyramid 3 。
Further, step 6 is described for LT 3 And LU 3 Carrying out weighted average fusion by adopting the ways of uncooled infrared image weight 1/4 and solar blind ultraviolet image weight 3/4 to obtain a fusion image LF 3 (ii) a For LT 2 And LU 2 、LT 1 And LU 1 Fusing in a way of increasing the gray value to obtain a fused image LF 2 And LF 1 The method comprises the following steps:
the fusion process is shown as formula (3):
wherein LF i Laplacian pyramid, LT, for fused images i Laplacian pyramid, LU as uncooled infrared image i Purple for the sun blindThe laplacian pyramid of the outer image.
Compared with the prior art, the invention has the following remarkable advantages: (1) the infrared image is imaged according to the thermal radiation of an object, is not interfered by the working period, can penetrate smoke at the same time, and can further judge the damage degree of the power transmission line after the ultraviolet detection determines a corona discharge point; (2) the fusion mode is simple in operation, few in resources required by algorithm implementation and good in image fusion effect.
Drawings
FIG. 1 is a flow chart of the non-refrigeration infrared image and solar blind ultraviolet image fusion algorithm based on tower decomposition.
Fig. 2 is an exploded view of the gaussian pyramid and laplacian pyramid of the uncooled infrared image of the present invention.
Fig. 3 is an exploded view of the gaussian pyramid and laplacian pyramid of the solar blind ultraviolet image of the present invention.
Fig. 4 is a diagram showing the fusion effect of the uncooled infrared image and the ultraviolet image in the present invention.
Detailed Description
The invention relates to a method for fusing an uncooled infrared image and a solar blind ultraviolet image based on tower decomposition, which comprises the following steps of:
acquiring an uncooled infrared image, firstly carrying out median filtering to remove salt and pepper noise, and then carrying out upper platform histogram equalization to improve the contrast of the whole image; respectively carrying out Gaussian pyramid decomposition and Laplacian pyramid decomposition on the uncooled infrared image and the ultraviolet image; in the fusion, the third layer of the Laplacian pyramid adopts weighted average fusion, and the second layer and the first layer adopt gray value large fusion; and reconstructing the three layers of Laplacian pyramids after layered fusion to obtain a final fusion image.
The invention relates to a tower decomposition-based method for fusing an uncooled infrared image and a solar blind ultraviolet image, which comprises the following steps of:
step 1, performing median filtering on an uncooled infrared image in a neighborhood range of 3 x 3 to remove salt and pepper noise;
step 2, performing upper platform histogram equalization on the non-refrigeration infrared image subjected to median filtering, and improving the contrast of the image;
step 3, taking the preprocessed uncooled infrared image as the first layer GT of the Gaussian pyramid 1 Using a Gaussian kernel g 3×3 To GT 1 Performing convolution and downsampling to remove even rows and columns to obtain a second layer GT of Gaussian pyramid 2 (ii) a With a Gaussian kernel g 3×3 To GT 2 Convolution and down-sampling are carried out to obtain a third layer GT of a Gaussian pyramid 3 ；
Step 4, for GT in step 3 2 Upsampling, even rows and columns filled with 0, and then using Gaussian kernel g 3×3 Performing convolution to obtain
To GT 3 Upsampling is performed, even rows and columns are filled with 0 s, and then with a Gaussian kernel g 3×3 Performing convolution to obtain
Use of
Obtaining a Laplacian pyramid first layer LT 1 Use of
Obtaining a Laplacian pyramid second layer LT 2 Laplacian pyramid third layer LT 3 And GT 3 The same;
and 5, repeating the step 3 and the step 4 for the solar blind ultraviolet image to obtain the Gaussian pyramid GU of the solar blind ultraviolet image i And Laplacian pyramid LU i ；
Step 6, for LT 3 And LU 3 Carrying out weighted average fusion by adopting the ways of uncooled infrared image weight 1/4 and solar blind ultraviolet image weight 3/4 to obtain a fusion image LF 3 (ii) a For LT 2 And LU 2 、LT 1 And LU 1 Fusing in a way of increasing the gray value to obtain a fused image LF 2 And LF 1 ；
Step 7, fusing images LF 3 Upsampling and using Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 2 Adding, up-sampling the resulting images, and applying a Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 1 And adding, and finishing reconstruction to obtain a final fusion image.
As a specific example, the Gaussian kernel g described in step 3 3×3 The method comprises the following steps:
as a specific example, the step 4 is described for GT in step 3 2 Upsampling, even rows and columns filled with 0, and then using Gaussian kernel g 3×3 Performing convolution to obtain
To GT 3 Upsampling is performed, even rows and columns are filled with 0 s, and then with a Gaussian kernel g 3×3 Performing convolution to obtain
Use of
Obtaining a Laplacian pyramid first layer LT 1 Use of
Obtaining a Laplacian pyramid second layer LT 2 Laplacian pyramid third layer LT 3 And GT 3 The same is true, specifically as follows:
the process of constructing the laplacian pyramid is shown in formula (2):
where UP represents an UP-sampling operation and,
denotes the convolution operation, g 3×3 Represents a gaussian kernel; third layer L of Laplacian pyramid 3 I.e. the third layer G of the Gaussian pyramid 3 。
As a specific example, the method described in step 6 for LT 3 And LU 3 Carrying out weighted average fusion by adopting the ways of uncooled infrared image weight 1/4 and solar blind ultraviolet image weight 3/4 to obtain a fusion image LF 3 (ii) a For LT 2 And LU 2 、LT 1 And LU 1 Fusing by adopting a mode of taking a large gray value to obtain a fused image LF 2 And LF 1 The method comprises the following steps:
the fusion process is shown as formula (3):
wherein LF i Laplacian pyramid, LT, for fused images i Laplacian pyramid, LU, as uncooled infrared images i Is a laplacian pyramid of a solar blind ultraviolet image.
The invention is described in further detail below with reference to the figures and the embodiments.
Examples
The embodiment is an uncooled infrared image and solar blind ultraviolet image fusion algorithm based on tower decomposition, and fully considers the characteristics of low background referential property and unclear details of an ultraviolet image, after the uncooled infrared image and the ultraviolet image are decomposed in the tower, the uncooled infrared image and the ultraviolet image are respectively fused in different spatial frequency intervals, so that the target characteristics in the ultraviolet image are guaranteed to be reserved, and the infrared image provides a background. Meanwhile, in order to reduce the operation amount, when the pixel level fusion is carried out, the weighted average fusion is carried out on the third layer of images, the gray value of the second layer of images and the gray value of the first layer of images are subjected to the large fusion, and the operation speed is improved.
As ultraviolet radiation less than 280nm in sunlight is absorbed by most parts when passing through an ozone layer in the atmosphere, a solar blind area is formed, and the imaging background is pure when ultraviolet radiation of 200-280 nm is detected near the earth surface. However, because the ultraviolet light emission source in the solar blind area wave band is not common, the solar blind ultraviolet imaging is often only provided with a signal at a target position, and specific background information cannot be distinguished in other areas. To provide background information for the ultraviolet image and aid in corona detection, the infrared image needs to be fused with the ultraviolet image.
Define a 3 × 3 Gaussian convolution kernel as g 3×3 Decomposition of the Gaussian pyramid of the uncooled infrared image into GT i ，GT i The image after upsampling and convolution with a Gaussian kernel is
The image of the Laplacian pyramid is LT i . For solar blind ultraviolet images, the images after decomposition of the Gaussian pyramids are GU i ，GU i The image after upsampling and convolution with a Gaussian kernel is
The image of the Laplacian pyramid is LU i . The Laplace pyramid image after the non-refrigeration infrared image and the solar blind ultraviolet image are fused is LF i ，LF i The images after upsampling and convolution with a gaussian kernel are
With reference to fig. 1, the method for fusing the uncooled infrared image and the solar blind ultraviolet image based on the tower decomposition in this embodiment includes the following specific steps:
step 1: performing median filtering on the uncooled infrared image in a neighborhood range of 3 x 3 to remove salt and pepper noise;
step 2: performing upper-platform histogram equalization on the non-refrigeration infrared image subjected to median filtering, and improving the contrast of the image;
and step 3: taking the preprocessed uncooled infrared image as the first layer GT of the Gaussian pyramid 1 Using a Gaussian kernel g 3×3 To GT 1 Performing convolution and downsampling to remove even rows and columns to obtain a second layer GT of Gaussian pyramid 2 (ii) a With a Gaussian kernel g 3×3 To GT 2 Convolution and down sampling are carried out to obtain a third layer GT of the Gaussian pyramid 3 ；
Further, the Gaussian core g 3×3 The method comprises the following steps:
and 4, step 4: for GT in step 3 2 Upsampling, even rows and columns filled with 0, and then using Gaussian kernel g 3×3 Performing convolution to obtain
To GT 3 Upsampling is performed, even rows and columns are filled with 0 s, and then with a Gaussian kernel g 3×3 Performing convolution to obtain
Use of
Obtaining a Laplacian pyramid first layer LT 1 Use of
To obtainLaplacian pyramid second layer LT 2 Laplacian pyramid third layer LT 3 And GT 3 The same;
further, the process of constructing the laplacian pyramid is specifically as follows:
where UP denotes an UP-sampling operation and,
denotes the convolution operation, g 3×3 Represents a gaussian kernel; third layer L of Laplacian pyramid 3 I.e. the third layer G of the Gaussian pyramid 3 . Fig. 2 is an exploded view of the gaussian pyramid and laplacian pyramid of the uncooled infrared image of the present invention.
And 5: and (4) repeating the step (3) and the step (4) for the solar blind ultraviolet image to obtain the Gaussian pyramid GU of the solar blind ultraviolet image i And Laplacian pyramid LU i (ii) a Fig. 3 is an exploded view of the gaussian pyramid and laplacian pyramid of the solar blind ultraviolet image of the present invention.
Step 6: for LT 3 And LU 3 Carrying out weighted average fusion by adopting the ways of uncooled infrared image weight 1/4 and solar blind ultraviolet image weight 3/4 to obtain a fusion image LF 3 (ii) a For LT 2 And LU 2 、LT 1 And LU 1 Fusing in a way of increasing the gray value to obtain a fused image LF 2 And LF 1 ；
Further, the fusion process is specifically as follows:
wherein LF i Laplacian pyramid, LT, for fused images i Laplacian pyramid, LU as uncooled infrared image i Is a laplacian pyramid of a solar blind ultraviolet image.
And 7: for the fusion image LF 3 Upsampling and using Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 2 Adding, up-sampling the resulting image, and applying a Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 1 And adding, and finishing reconstruction to obtain a final fusion image. Fig. 4 is a diagram showing the fusion effect of the uncooled infrared image and the ultraviolet image in the present invention.
In conclusion, the fusion method of the invention has simple operation, less resources required by the algorithm and good image fusion effect.
Claims (3)
1. A method for fusing an uncooled infrared image and a solar blind ultraviolet image based on tower decomposition is characterized by comprising the following steps:
acquiring an uncooled infrared image, firstly carrying out median filtering to remove salt and pepper noise, and then carrying out upper platform histogram equalization to improve the contrast of the whole image; respectively carrying out Gaussian pyramid decomposition and Laplacian pyramid decomposition on the uncooled infrared image and the ultraviolet image; in the fusion, the Laplace pyramid third layer adopts weighted average fusion, and the second layer and the first layer adopt gray value taking large fusion; reconstructing the three layers of Laplacian pyramids after hierarchical fusion to obtain a final fusion image;
the non-refrigeration infrared image and solar blind ultraviolet image fusion method based on tower decomposition comprises the following steps:
step 1, performing median filtering on the uncooled infrared image to remove salt and pepper noise;
step 2, performing upper platform histogram equalization on the non-refrigeration infrared image subjected to median filtering;
step 3, taking the preprocessed uncooled infrared image as a first layer GT of a Gaussian pyramid 1 Using a Gaussian kernel g 3×3 To GT 1 Performing convolution and downsampling to remove even rows and columns to obtain a second layer GT of Gaussian pyramid 2 (ii) a By using a Gaussian kernel g 3×3 To GT 2 Convolution and down sampling are carried out to obtain a third layer GT of the Gaussian pyramid 3 ；
Step 4, for GT in step 3 2 Upsampled, even rows and columns filled with 0, and then with a Gaussian kernel g 3×3 Performing convolution to obtain
To GT 3 Upsampling is performed, even rows and columns are filled with 0 s, and then with a Gaussian kernel g 3×3 Performing convolution to obtain
Use of
Obtaining a Laplacian pyramid first layer LT 1 Use of
Obtaining a Laplacian pyramid second layer LT 2 Laplacian pyramid third layer LT 3 And GT 3 The same is true, specifically as follows:
the process of constructing the laplacian pyramid is shown in formula (2):
where UP represents an UP-sampling operation and,
denotes the convolution operation, g 3×3 Represents a gaussian kernel; third layer L of Laplacian pyramid 3 I.e. the third layer G of the Gaussian pyramid 3 ；
And 5, repeating the step 3 and the step 4 for the solar blind ultraviolet image to obtain the Gaussian pyramid GU of the solar blind ultraviolet image i And Laplacian pyramid LU i ；
Step 6, for LT 3 And LU 3 Carrying out weighted average fusion by adopting the ways of uncooled infrared image weight 1/4 and solar blind ultraviolet image weight 3/4 to obtain a fusion image LF 3 (ii) a For LT 2 And LU 2 、LT 1 And LU 1 Fusing in a way of increasing the gray value to obtain a fused image LF 2 And LF 1 ；
Step 7, fusing images LF 3 Upsampling and using Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 2 Adding, up-sampling the resulting images, and applying a Gaussian kernel g 3×3 Performing convolution to obtain an image
Will be provided with
And LF 1 And adding, and finishing reconstruction to obtain a final fusion image.
2. The method for fusing the uncooled infrared image and the solar-blind ultraviolet image based on the tower decomposition as claimed in claim 1, wherein the Gaussian kernel g in the step 3 3×3 The method comprises the following steps:
3. the method for fusing uncooled infrared image and solar-blind ultraviolet image based on tower decomposition as claimed in claim 1, wherein the step 6 is for LT 3 And LU 3 Carrying out weighted average fusion by adopting the ways of uncooled infrared image weight 1/4 and solar blind ultraviolet image weight 3/4 to obtain a fusion image LF 3 (ii) a For LT 2 And LU 2 、LT 1 And LU 1 Fusing by adopting a mode of taking a large gray value to obtain a fused image LF 2 And LF 1 The method comprises the following steps:
the fusion process is shown as formula (3):
wherein LF i Laplacian pyramid, LT, for fused images i Laplacian pyramid, LU as uncooled infrared image i Is a laplacian pyramid of a solar blind ultraviolet image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811586175.XA CN109816616B (en) | 2018-12-25 | 2018-12-25 | Non-refrigeration infrared image and solar blind ultraviolet image fusion method based on tower decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811586175.XA CN109816616B (en) | 2018-12-25 | 2018-12-25 | Non-refrigeration infrared image and solar blind ultraviolet image fusion method based on tower decomposition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109816616A CN109816616A (en) | 2019-05-28 |
| CN109816616B true CN109816616B (en) | 2022-09-06 |
Family
ID=66602376
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811586175.XA Active CN109816616B (en) | 2018-12-25 | 2018-12-25 | Non-refrigeration infrared image and solar blind ultraviolet image fusion method based on tower decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109816616B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111462032B (en) * | 2020-03-31 | 2023-03-31 | 北方夜视技术股份有限公司 | Method for fusing uncooled infrared image and solar blind ultraviolet image and application |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103578091A (en) * | 2013-10-10 | 2014-02-12 | 中国科学院上海技术物理研究所 | Infrared polarization image fusion method based on Laplacian pyramid |
| CN106570831A (en) * | 2016-10-09 | 2017-04-19 | 中国航空工业集团公司洛阳电光设备研究所 | Gray image contrast equalization enhancement method |
-
2018
- 2018-12-25 CN CN201811586175.XA patent/CN109816616B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103578091A (en) * | 2013-10-10 | 2014-02-12 | 中国科学院上海技术物理研究所 | Infrared polarization image fusion method based on Laplacian pyramid |
| CN106570831A (en) * | 2016-10-09 | 2017-04-19 | 中国航空工业集团公司洛阳电光设备研究所 | Gray image contrast equalization enhancement method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109816616A (en) | 2019-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107145846B (en) | A kind of insulator recognition methods based on deep learning | |
| US10235746B2 (en) | Method and device for fusing panchromatic image and infrared image | |
| CN108596103B (en) | High-resolution remote sensing image building extraction method based on optimal spectral index selection | |
| CN103049898B (en) | Method for fusing multispectral and full-color images with light cloud | |
| CN106877237A (en) | A kind of method of insulator missing in detection transmission line of electricity based on Aerial Images | |
| CN107784642B (en) | A kind of infrared video and visible light video method for self-adaption amalgamation | |
| CN110189286B (en) | Infrared and visible light image fusion method based on ResNet | |
| CN103700075B (en) | Multichannel satellite cloud picture fusion method based on Tetrolet conversion | |
| CN103295204B (en) | A kind of image self-adapting enhancement method based on non-down sampling contourlet transform | |
| CN102163329A (en) | Super-resolution reconstruction method of single-width infrared image based on scale analogy | |
| CN109816616B (en) | Non-refrigeration infrared image and solar blind ultraviolet image fusion method based on tower decomposition | |
| CN104484859A (en) | Multispectral optical remote sensing image data thin-cloud removing method | |
| CN107169947A (en) | A kind of image co-registration experimental method of feature based point location and rim detection | |
| CN113793289B (en) | Multispectral image and full-color image fuzzy fusion method based on CNN and NSCT | |
| CN112666426A (en) | Equipment fault detection method and system based on multispectral three-in-one image | |
| Lin et al. | Adaptive infrared and visible image fusion method by using rolling guidance filter and saliency detection | |
| CN112215787A (en) | Infrared and visible light image fusion method based on significance analysis and adaptive filter | |
| CN103049889A (en) | Preprocessing method for strong noise aero-optic effect degraded image | |
| CN112669201B (en) | Visible light cloud image conversion method and system based on infrared light and terminal thereof | |
| CN104240208A (en) | Uncooled infrared focal plane detector image detail enhancement method | |
| CN103530860B (en) | Adaptive autoregressive model-based hyper-spectral imagery super-resolution method | |
| CN104166975A (en) | Low-altitude infrared target detection algorithm in complex environment | |
| CN110175959B (en) | Typhoon cloud picture enhancement method | |
| Selvaraj et al. | Infrared and visible image fusion using multi‐scale NSCT and rolling‐guidance filter | |
| CN115294001A (en) | Night light remote sensing image fusion method for improving IHS and wavelet transformation |
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 |