CN116993626A - Infrared image noise reduction method and system based on time-space domain - Google Patents
Infrared image noise reduction method and system based on time-space domain Download PDFInfo
- Publication number
- CN116993626A CN116993626A CN202311243492.2A CN202311243492A CN116993626A CN 116993626 A CN116993626 A CN 116993626A CN 202311243492 A CN202311243492 A CN 202311243492A CN 116993626 A CN116993626 A CN 116993626A
- Authority
- CN
- China
- Prior art keywords
- image
- patch
- map
- pixel point
- weights
- 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.)
- Pending
Links
- 230000009467 reduction Effects 0.000 title claims abstract description 31
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000010606 normalization Methods 0.000 claims abstract description 16
- 238000009825 accumulation Methods 0.000 claims description 34
- HXGZBJUOISOSHJ-BFHYXJOUSA-N (4-aminophenyl) [(2r,3s,5r)-2-(hydroxymethyl)-5-(5-methyl-2,4-dioxopyrimidin-1-yl)oxolan-3-yl] hydrogen phosphate Chemical compound O=C1NC(=O)C(C)=CN1[C@@H]1O[C@H](CO)[C@@H](OP(O)(=O)OC=2C=CC(N)=CC=2)C1 HXGZBJUOISOSHJ-BFHYXJOUSA-N 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000004927 fusion Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 14
- 238000007499 fusion processing Methods 0.000 abstract description 3
- 238000001914 filtration Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 3
- 238000009499 grossing Methods 0.000 description 3
- 235000002566 Capsicum Nutrition 0.000 description 1
- 208000003351 Melanosis Diseases 0.000 description 1
- 239000006002 Pepper Substances 0.000 description 1
- 235000016761 Piper aduncum Nutrition 0.000 description 1
- 235000017804 Piper guineense Nutrition 0.000 description 1
- 244000203593 Piper nigrum Species 0.000 description 1
- 235000008184 Piper nigrum Nutrition 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004297 night vision Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 235000012434 pretzels Nutrition 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
- G06T7/41—Analysis of texture based on statistical description of texture
-
- 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
- G06T2207/10048—Infrared image
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a time-space domain-based infrared image noise reduction method and a system, wherein the method comprises the following steps: s1: taking the first pixel point of the image as a pixel point to be processed; s2: traversing pixel points to be processed to obtain an image block; s3: calculating the weight of each image block; s4: weighting and fusing the weights of all the image blocks; s5: judging whether the pixel point to be processed is the last pixel point, if so, executing a step S6; if not, returning to the step S2; s6: and carrying out normalization processing to obtain a denoising image dst. The application can not only utilize the similarity of the center points of the similar blocks in the fusion process, but also utilize the similarity among other pixel points of the similar blocks, thereby improving the effect of removing image noise on the basis of excessive loss of image details.
Description
Technical Field
The application relates to the technical field of infrared image processing, in particular to a time-space domain-based infrared image noise reduction method and system.
Background
The infrared imaging technology can still generate clear images in a dim environment and is widely applied to the fields of night vision devices, unmanned aerial vehicles, thermal imaging and the like. However, due to the complexity of the imaging environment, various noises often exist in the infrared image, and these noises seriously affect the quality of the infrared image and the accuracy of information. Therefore, researchers at home and abroad develop a series of research works around infrared image noise reduction. Currently, the infrared image noise reduction technology can be classified into 2D noise reduction and 3D noise reduction.
2D noise reduction, a common 2D noise reduction algorithm is as follows:
(1) The basic idea of the algorithm is to use a gaussian kernel to convolve an image, and to perform weighted average on pixels closer to the current pixel, so as to obtain a smoothed pixel value. The algorithm has a good effect of removing Gaussian noise in the image, but also loses important detail information in some images.
(2) The average filtering algorithm is a linear smoothing algorithm, and the basic idea is to replace the current pixel value with the average value of the neighborhood pixels around each pixel, so as to realize the noise reduction effect. However, the algorithm is sensitive to detail information such as edges and textures, and important information can be lost.
(3) The median filtering algorithm is a nonlinear smoothing algorithm, and the basic idea is to replace the current pixel value with the median of the neighborhood pixel gray values, so that large-aperture noise such as pretzel noise can be effectively removed, but the denoising effect on small-aperture noise such as Gaussian noise is poor.
(4) The double-sided filtering algorithm is a weighted smoothing algorithm based on the distance between pixels and gray level difference, and can remove noise and retain image edge information. The basic idea of the algorithm is to perform weighted average on the local neighborhood of a certain pixel under different spatial scale and gray scale conditions.
(5) NLM (Non-LocalMeans) image noise reduction algorithm is a nonlinear filtering algorithm based on image self-similarity. The method and the device utilize the similarity between pixels in the image to reduce noise, and can effectively remove noise while keeping image details.
The basic idea of the NLM algorithm is: and calculating a neighborhood window of each pixel point, and then carrying out similarity matching on the neighborhood windows. For the pixel to be processed, a weighted average method is used to calculate its output value. The weight of each neighborhood in the weighted average depends not only on its distance from the pixel point to be processed, but also on the gray-scale based similarity between pixels in the neighborhood window.
(6) The BM3D, BM3D (Block-Matching 3D) image denoising algorithm is a 3D filtering algorithm based on Block Matching, and can process image noise while keeping details of images, so that the processing effect is very good. The BM3D algorithm is a relatively typical image noise reduction algorithm, and has been widely used in the field of image processing. The BM3D algorithm can effectively remove noise, can keep detailed information of images, and has good effect when processing various types of noise. However, it is computationally intensive and requires a large amount of memory space, so that it is necessary to use acceleration techniques or GPU-based parallel computations for optimization when processing large images.
Two, 3D noise reduction, common 3D noise reduction algorithms are as follows:
(1) Based on block matching, the algorithm based on block matching utilizes the similarity among image blocks to perform block matching and 3D filtering on 3D volume data, so that noise such as salt and pepper noise, gaussian noise and the like can be removed, meanwhile, detailed information in the volume data is reserved, and the classical algorithm comprises VNLM, VBM3D, VBM D and the like.
(2) Based on motion estimation, the motion trail of the pixel points in the time neighborhood is obtained by establishing a motion model, and then time domain filtering is carried out according to the motion trail, wherein typical algorithms include 3D-NLM+, burst, meshflow and the like.
Compared with a 2D image, the 3D noise reduction can utilize more abundant image information, so that the noise reduction effect is often stronger.
The following technical problems exist in the prior art: the existing infrared image noise reduction technology is difficult to consider the denoising effect and detail reservation, the detail is seriously lost due to the strong denoising effect, and the detail reservation causes poor denoising effect (residual freckle block processing trace).
Disclosure of Invention
The application aims to provide a time-space domain-based infrared image noise reduction method and system, which are used for solving the technical problem of how to improve the effect of removing image noise on the basis of retaining image details.
The application is realized by adopting the following technical scheme: an infrared image noise reduction method based on a time-space domain comprises the following steps:
s1: taking the first pixel point of the image as a pixel point to be processed;
s2: traversing pixel points to be processed to obtain an image block;
s3: calculating the weight of each image block;
s4: weighting and fusing the weights of all the image blocks;
s5: judging whether the pixel point to be processed is the last pixel point, if so, executing a step S6; if not, returning to the step S2;
s6: and carrying out normalization processing to obtain a denoising image dst.
Further, the step S1 specifically includes: and initializing a pixel value accumulation map dst_map_tmp and a weight accumulation map weights_map which have the same width and height as the original image by taking the first pixel point of the image as a pixel point to be processed.
Further, step S2 includes the following sub-steps:
s21: taking the current pixel point as a center, and marking a square area with a side length d in the current frame image as a reference block, and marking the square area as a patch_i;
s22: at the same position of the current frame and the historical frame, a square area with the side length of D is defined as a search range;
s23: and traversing each pixel point in the search range to obtain a plurality of image blocks with the same size as the patch_i, and recording the image blocks as the patch_j.
Further, step S23 specifically includes: traversing each pixel point in the search range to obtain a common valueThe image blocks with the same size as the patch_i are marked as patch_j; if the image is n frames, can obtainImage blocks.
Further, the step S3 specifically includes: calculating the weight of each patch_j except for the patch_i, wherein the weight of each patch_j is determined by the similarity between the patch_j and the patch_i, and the calculation formula is as follows:
；
wherein ,for the weights of the image blocks +.>Representing the value of the kth element in block patch i,/and>representing the value of the kth element in block patch j,/and>is the total number of pixels of block patch_i, < >>Is a Gaussian window width, represents the decay rate of the weight with distance, and dist (i, j) represents the difference between patch_i and patch_j; the weight of patch_i is equal to the maximum of all patch_j weights.
Further, the step S4 specifically includes: the weights multiplied by the pixel values in the blocks of the latch_i and the latch_j are added to the pixel value accumulation map dst_map_tmp at the same position as the latch_i, and the weights of the image blocks are added to the weight accumulation map weights_map at the same position as the latch_i.
Further, the step S6 specifically includes: dividing the pixel value accumulation map dst_map_tmp point by the weight accumulation map weights_map, and carrying out normalization to obtain a final denoising image dst, wherein the calculation formula is as follows:
。
an infrared image noise reduction system based on a time-space domain comprises an initialization module, an image block calculation module, an image block processing module, a judgment module and a normalization module, wherein,
the initialization module is used for taking the first pixel point of the image as a pixel point to be processed, and initializing a pixel value accumulation graph dst_map_tmp and a weight accumulation graph weights_map which have the same width and height as the original image;
the image block calculation module is used for traversing the pixel points to be processed to obtain an image block;
the image block processing module is used for calculating the weight of each image block and carrying out weighted fusion on the weights of all the image blocks;
the judging module is used for judging whether the pixel point to be processed is the last pixel point;
the normalization module is used for performing normalization processing on the pixel value accumulation graph dst_map_tmp and the weight accumulation graph weights_map to obtain a denoising image dst.
The application has the beneficial effects that: compared with the existing algorithm based on 3D block matching, the method can utilize the similarity of the center points of the similar blocks in the fusion process, and can also utilize the similarity among other pixel points of the similar blocks, so that the effect of removing image noise can be improved on the basis of excessive loss of image details.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the present application;
FIG. 2 is a schematic diagram of traversing a search range;
FIG. 3 is a weighted fusion schematic;
FIG. 4 is a diagram of an example of pixel value accumulation;
fig. 5 is a diagram of an example of weight accumulation.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
Some embodiments of the present application are described in detail below with reference to the accompanying drawings. The following embodiments and features of the embodiments may be combined with each other without conflict.
Example 1
Referring to fig. 1, a time-space domain based infrared image denoising method includes the following steps:
s1: taking the first pixel point of the image as a pixel point to be processed;
s2: traversing pixel points to be processed to obtain an image block;
s3: calculating the weight of each image block;
s4: weighting and fusing the weights of all the image blocks;
s5: judging whether the pixel point to be processed is the last pixel point, if so, executing a step S6; if not, returning to the step S2;
s6: and carrying out normalization processing to obtain a denoising image dst.
The step S1 specifically comprises the following steps: and initializing a pixel value accumulation graph dst_map_tmp and a weight accumulation graph weights_map with the same width and height as the original graph by taking the first pixel point of the image as a pixel point to be processed, wherein the initial values are 0.
Referring to fig. 2, step S2 specifically includes: and taking the current pixel point as a center, marking a square area with the side length of D in the current frame image as a reference block, marking the square area as a patch_i, and marking the square area with the side length of D as a search range at the same position of the current frame and the historical frame. Traversing each pixel point in the search range to obtain a common valueThe image blocks with the same size as the patch_i are marked as patch_j, and the total of n frames of images is +.>Image blocks.
The step S3 specifically comprises the following steps: calculating the weight of each patch_j except for the patch_i, wherein the weight of each patch_j is determined by the similarity between the patch_j and the patch_i, and the detailed calculation formula is as follows:
； wherein ,/>For the weights of the image blocks +.>Representing the value of the kth element in block patch i,/and>representing the value of the kth element in block patch j,/and>the total number of pixels of the block patch_i, dist (i, j) represents the difference between the patch_i and the patch_j (the pixel values at the corresponding positions are subjected to difference, the square sum of all the differences is obtained, and the sum of the differences is divided by the total number of pixel points in the block for normalization); />Is a gaussian window width, represents the decay rate of the weight with distance,can be adjusted according to actual conditions. Wherein the weight of patch_i is equal to the maximum of all patch_j weights.
Referring to fig. 3, step S4 specifically includes: the weights multiplied by the pixel values in the blocks of the latch_i and the latch_j are added to the pixel value accumulation map dst_map_tmp at the same position as the latch_i, and the weights of the image blocks are added to the weight accumulation map weights_map at the same position as the latch_i. All image blocks are multiplied by the respective weights and added to the same location as the center block. A specific method of pixel value accumulation is shown in fig. 4, in which,
。
for the purpose of this description, assuming that the block size is 3*3, there are only 3 blocks in the search range, and the pixel value A1 of the first pixel point in the accumulated result is equal to the pixel value of the pixel point at the corresponding position of each block multiplied by the weight of the block where the pixel point is located, and then summed. The specific method of weight accumulation is also similar, as shown in fig. 5, wherein,
。
the step S5 specifically comprises the following steps: and judging whether the pixel point to be processed is the last pixel point, if so, executing the step S6, otherwise, updating the pixel point to be processed as the next point, and returning to the step S2.
The step S6 specifically comprises the following steps: the final denoised image dst is obtained by normalizing the pixel value accumulation map dst_map_tmp point division weight accumulation map weight_map, and the calculation formula is as follows:
。
an infrared image noise reduction system based on a time-space domain comprises an initialization module, an image block calculation module, an image block processing module, a judgment module and a normalization module, wherein,
the initialization module is used for taking the first pixel point of the image as a pixel point to be processed, and initializing a pixel value accumulation graph dst_map_tmp and a weight accumulation graph weights_map which have the same width and height as the original image;
the image block calculation module is used for traversing the pixel points to be processed to obtain an image block;
the image block processing module is used for calculating the weight of each image block and carrying out weighted fusion on the weights of all the image blocks;
the judging module is used for judging whether the pixel point to be processed is the last pixel point;
the normalization module is used for performing normalization processing on the pixel value accumulation graph dst_map_tmp and the weight accumulation graph weights_map to obtain a denoising image dst.
Based on the above embodiments, the present application has at least the following technical effects:
compared with the existing algorithm based on 3D block matching, the method can utilize the similarity of the center points of the similar blocks in the fusion process, and can also utilize the similarity among other pixel points of the similar blocks, so that the effect of removing image noise can be improved on the basis of excessive loss of image details.
It should be noted that, for simplicity of description, the foregoing embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, it should be understood by those skilled in the art that the embodiments described in the specification are preferred embodiments and that the actions involved are not necessarily required for the present application.
In the above embodiments, the basic principle and main features of the present application and advantages of the present application are described. It will be appreciated by persons skilled in the art that the present application is not limited by the foregoing embodiments, but rather is shown and described in what is considered to be illustrative of the principles of the application, and that modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the application, and therefore, is within the scope of the appended claims.
Claims (8)
1. The infrared image noise reduction method based on the time-space domain is characterized by comprising the following steps of:
s1: taking the first pixel point of the image as a pixel point to be processed;
s2: traversing pixel points to be processed to obtain an image block;
s3: calculating the weight of each image block;
s4: weighting and fusing the weights of all the image blocks;
s5: judging whether the pixel point to be processed is the last pixel point, if so, executing a step S6; if not, returning to the step S2;
s6: and carrying out normalization processing to obtain a denoising image dst.
2. The method for noise reduction of an infrared image based on a time-space domain as set forth in claim 1, wherein step S1 specifically includes: and initializing a pixel value accumulation map dst_map_tmp and a weight accumulation map weights_map which have the same width and height as the original image by taking the first pixel point of the image as a pixel point to be processed.
3. The method for noise reduction of an infrared image based on a time-space domain as set forth in claim 2, wherein the step S2 comprises the sub-steps of:
s21: taking the current pixel point as a center, and marking a square area with a side length d in the current frame image as a reference block, and marking the square area as a patch_i;
s22: at the same position of the current frame and the historical frame, a square area with the side length of D is defined as a search range;
s23: and traversing each pixel point in the search range to obtain a plurality of image blocks with the same size as the patch_i, and recording the image blocks as the patch_j.
4. The method for noise reduction of an infrared image based on a time-space domain as set forth in claim 3, wherein the step S23 specifically includes: traversing each pixel point in the search range to obtain a common valueThe image blocks with the same size as the patch_i are marked as patch_j; if n frames of images are available +.>Image blocks.
5. The method for noise reduction of an infrared image based on a time-space domain as set forth in claim 3, wherein the step S3 is specifically: calculating the weight of each patch_j except for the patch_i, wherein the weight of each patch_j is determined by the similarity between the patch_j and the patch_i, and the calculation formula is as follows:
；
wherein ,for the weights of the image blocks +.>Representing the value of the kth element in block patch i,/and>representing the value of the kth element in block patch j,/and>is the total number of pixels of block patch_i, < >>Is a Gaussian window width, represents the decay rate of the weight with distance, and dist (i, j) represents the difference between patch_i and patch_j; the weight of patch_i is equal to the maximum of all patch_j weights.
6. The method for noise reduction of an infrared image based on a time-space domain as set forth in claim 5, wherein step S4 specifically includes: the weights multiplied by the pixel values in the blocks of the latch_i and the latch_j are added to the pixel value accumulation map dst_map_tmp at the same position as the latch_i, and the weights of the image blocks are added to the weight accumulation map weights_map at the same position as the latch_i.
7. The method for noise reduction of an infrared image based on a time-space domain as set forth in claim 6, wherein step S6 is specifically: dividing the pixel value accumulation map dst_map_tmp point by the weight accumulation map weights_map, and carrying out normalization to obtain a final denoising image dst, wherein the calculation formula is as follows:
。
8. an infrared image noise reduction system based on a time-space domain is used for realizing the infrared image noise reduction method based on the time-space domain as set forth in any one of claims 1-7, and is characterized by comprising an initialization module, an image block calculation module, an image block processing module, a judgment module and a normalization module, wherein,
the initialization module is used for taking the first pixel point of the image as a pixel point to be processed, and initializing a pixel value accumulation graph dst_map_tmp and a weight accumulation graph weights_map which have the same width and height as the original image;
the image block calculation module is used for traversing the pixel points to be processed to obtain an image block;
the image block processing module is used for calculating the weight of each image block and carrying out weighted fusion on the weights of all the image blocks;
the judging module is used for judging whether the pixel point to be processed is the last pixel point;
the normalization module is used for performing normalization processing on the pixel value accumulation graph dst_map_tmp and the weight accumulation graph weights_map to obtain a denoising image dst.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311243492.2A CN116993626A (en) | 2023-09-26 | 2023-09-26 | Infrared image noise reduction method and system based on time-space domain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311243492.2A CN116993626A (en) | 2023-09-26 | 2023-09-26 | Infrared image noise reduction method and system based on time-space domain |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116993626A true CN116993626A (en) | 2023-11-03 |
Family
ID=88521665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311243492.2A Pending CN116993626A (en) | 2023-09-26 | 2023-09-26 | Infrared image noise reduction method and system based on time-space domain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116993626A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102005033A (en) * | 2010-11-16 | 2011-04-06 | 中国科学院遥感应用研究所 | Method for suppressing noise by image smoothing |
| CN102136134A (en) * | 2011-03-11 | 2011-07-27 | 西安电子科技大学 | Method for removing speckles from SAR (Synthetic Aperture Radar) image based on MRF (Markov Random Field) priority |
| CN107025640A (en) * | 2017-04-28 | 2017-08-08 | 南京觅踪电子科技有限公司 | The Medical Image Denoising method of non-local mean based on optimization |
| CN107818547A (en) * | 2016-09-14 | 2018-03-20 | 北京航空航天大学 | The minimizing technology of the spiced salt and Gaussian mixed noise in a kind of sequence towards twilight image |
| CN109859126A (en) * | 2019-01-17 | 2019-06-07 | 浙江大华技术股份有限公司 | A kind of vedio noise reduction method, apparatus, electronic equipment and storage medium |
| CN110648290A (en) * | 2019-09-06 | 2020-01-03 | 西安交通大学 | Dual-core non-local mean image denoising method based on SURE parameter optimization |
| CN110910324A (en) * | 2019-11-19 | 2020-03-24 | 山东神戎电子股份有限公司 | Method for removing vertical stripes of infrared video |
| CN113239860A (en) * | 2021-05-28 | 2021-08-10 | 中建材信息技术股份有限公司 | Smoke and fire detection method based on video |
| CN113438386A (en) * | 2021-05-20 | 2021-09-24 | 珠海全志科技股份有限公司 | Dynamic and static judgment method and device applied to video processing |
| CN115578287A (en) * | 2022-10-31 | 2023-01-06 | 浙江大华技术股份有限公司 | Image noise reduction method, device, equipment and storage medium |
| CN116016807A (en) * | 2022-12-30 | 2023-04-25 | 广东中星电子有限公司 | Video processing method, system, storable medium and electronic equipment |
-
2023
- 2023-09-26 CN CN202311243492.2A patent/CN116993626A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102005033A (en) * | 2010-11-16 | 2011-04-06 | 中国科学院遥感应用研究所 | Method for suppressing noise by image smoothing |
| CN102136134A (en) * | 2011-03-11 | 2011-07-27 | 西安电子科技大学 | Method for removing speckles from SAR (Synthetic Aperture Radar) image based on MRF (Markov Random Field) priority |
| CN107818547A (en) * | 2016-09-14 | 2018-03-20 | 北京航空航天大学 | The minimizing technology of the spiced salt and Gaussian mixed noise in a kind of sequence towards twilight image |
| CN107025640A (en) * | 2017-04-28 | 2017-08-08 | 南京觅踪电子科技有限公司 | The Medical Image Denoising method of non-local mean based on optimization |
| CN109859126A (en) * | 2019-01-17 | 2019-06-07 | 浙江大华技术股份有限公司 | A kind of vedio noise reduction method, apparatus, electronic equipment and storage medium |
| CN110648290A (en) * | 2019-09-06 | 2020-01-03 | 西安交通大学 | Dual-core non-local mean image denoising method based on SURE parameter optimization |
| CN110910324A (en) * | 2019-11-19 | 2020-03-24 | 山东神戎电子股份有限公司 | Method for removing vertical stripes of infrared video |
| CN113438386A (en) * | 2021-05-20 | 2021-09-24 | 珠海全志科技股份有限公司 | Dynamic and static judgment method and device applied to video processing |
| CN113239860A (en) * | 2021-05-28 | 2021-08-10 | 中建材信息技术股份有限公司 | Smoke and fire detection method based on video |
| CN115578287A (en) * | 2022-10-31 | 2023-01-06 | 浙江大华技术股份有限公司 | Image noise reduction method, device, equipment and storage medium |
| CN116016807A (en) * | 2022-12-30 | 2023-04-25 | 广东中星电子有限公司 | Video processing method, system, storable medium and electronic equipment |
Non-Patent Citations (3)
| Title |
|---|
| 何艳敏 等: "基于稀疏表示的图像压缩和去噪理论与应用", 电子科技大学出版社, pages: 102 - 107 * |
| 杨婧: "基于时空域去噪算法的视频处理模块设计", 中国优秀硕士学位论文全文数据库 信息科技辑, pages 138 - 1098 * |
| 游青艳: "基于时空域的网络监控视频去噪研究", 中国优秀硕士学位论文全文数据库 信息科技辑, pages 138 - 1175 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107369166B (en) | Target tracking method and system based on multi-resolution neural network | |
| CN102768760B (en) | Quick image dehazing method on basis of image textures | |
| Obradović et al. | Linear fuzzy space based road lane model and detection | |
| CN101551901B (en) | Method for compensating and enhancing dynamic shielded image in real time | |
| CN104463819B (en) | Image filtering method and device | |
| CN105184743B (en) | A kind of image enchancing method based on non-linear Steerable filter | |
| CN105046659A (en) | Sparse representation-based single lens calculation imaging PSF estimation method | |
| CN105913392A (en) | Degraded image overall quality improving method in complex environment | |
| CN102968770A (en) | Method and device for eliminating noise | |
| CN102567973A (en) | Image denoising method based on improved shape self-adaptive window | |
| CN111899200B (en) | Infrared image enhancement method based on 3D filtering | |
| CN110796616A (en) | Fractional order differential operator based L0Norm constraint and adaptive weighted gradient turbulence degradation image recovery method | |
| CN109064402B (en) | Single image super-resolution reconstruction method based on enhanced non-local total variation model prior | |
| CN107451986B (en) | Single infrared image enhancement method based on fusion technology | |
| CN102222327A (en) | Image denoising method based on Treelet transformation and minimum mean-square error estimation | |
| CN111931722A (en) | Correlated filtering tracking method combining color ratio characteristics | |
| CN102136134B (en) | Method for removing speckles from SAR (Synthetic Aperture Radar) image based on MRF (Markov Random Field) priority | |
| CN109658357A (en) | A kind of denoising method towards remote sensing satellite image | |
| CN113177898A (en) | Image defogging method and device, electronic equipment and storage medium | |
| CN105303544A (en) | Video splicing method based on minimum boundary distance | |
| CN109242797A (en) | Image de-noising method, system and the medium merged based on homogeneous and heterogeneous areas | |
| CN116012265B (en) | Infrared video denoising method and device based on time-space domain adaptive filtering | |
| CN116993626A (en) | Infrared image noise reduction method and system based on time-space domain | |
| CN116862809A (en) | Image enhancement method under low exposure condition | |
| Soumya et al. | Self-organized night video enhancement for surveillance systems |
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 |