CN116012265B - Infrared video denoising method and device based on time-space domain adaptive filtering - Google Patents
Infrared video denoising method and device based on time-space domain adaptive filtering Download PDFInfo
- Publication number
- CN116012265B CN116012265B CN202310312205.2A CN202310312205A CN116012265B CN 116012265 B CN116012265 B CN 116012265B CN 202310312205 A CN202310312205 A CN 202310312205A CN 116012265 B CN116012265 B CN 116012265B
- Authority
- CN
- China
- Prior art keywords
- image
- infrared video
- tailing
- moving object
- time
- 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
- 238000001914 filtration Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 23
- 239000000523 sample Substances 0.000 claims abstract description 33
- 230000009467 reduction Effects 0.000 claims abstract description 28
- 238000013135 deep learning Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 12
- 238000003709 image segmentation Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 3
- 230000002146 bilateral effect Effects 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 239000004566 building material Substances 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The infrared video noise reduction method based on the time-space domain adaptive filtering is used for removing the tail of an indoor moving object, and the moving object is segmented according to a mask to obtain a moving image of the moving object; judging a fuzzy region of a moving image to obtain a tailing image, performing deep learning on the moving image by using a memory module, performing super-resolution three-dimensional reconstruction on the moving image by using a deep learning algorithm to obtain three-dimensional data of a moving object, and obtaining a tailing reconstructed image in the three-dimensional data by using calibration information of a probe; and (3) utilizing a tailing probe to position, and superposing a tailing reconstructed image to the area of the tailing probe of the input infrared video image to realize tailing removal of the moving object. According to the invention, the smear region and the three-dimensional stereo data are mapped through the probe of the real-time smear image, and the smear reconstruction image corresponding to the smear region is obtained, so that the smear removal calculated amount of the moving object is reduced, and the removal speed is increased.
Description
Technical Field
The invention relates to the technical field of imaging of infrared images, in particular to an infrared video noise reduction method and device based on time-space domain adaptive filtering.
Background
The infrared image is the product of the combination of infrared technology and imaging technology, which can be used for temperature measurement, fire monitoring, military, building material detection, medicine, power industry and the like. But the infrared thermal imaging image has the characteristics of lower resolution, fuzzy details and poor adaptability to fast moving targets.
This is due to the imaging mechanism of the infrared image and the infrared imaging system itself, and there has been no effective solution to the problem of moving object tailing and blurring in the infrared video image.
Accordingly, the problems of the prior art are to be further improved and developed.
Disclosure of Invention
(one) object of the invention: in order to solve the problems in the prior art, the invention aims to provide a noise reduction method and a noise reduction device capable of efficiently removing the tail of an indoor moving object so as to improve the definition of an infrared video image.
(II) technical scheme: in order to solve the technical problems, the technical scheme provides an infrared video noise reduction method based on time-space domain adaptive filtering, which is used for removing the tail of an indoor moving object and comprises the following steps:
step A, obtaining pixels of an input infrared video image, calculating a local standard image and a local mean value image of the pixels in the input infrared video image, and calculating a local gray scale fluctuation rate image of the pixels in the input infrared video image according to the local standard image and the local mean value image;
b, performing soft segmentation on an input infrared video image according to the local gray scale fluctuation rate image to obtain a mask of a detail area, and performing moving object segmentation on the mask to obtain a moving image of a moving object;
step C, judging a blurring region of the moving image to obtain a tailing image, performing deep learning on the moving image by using a memory module, performing super-resolution three-dimensional reconstruction on the moving image by using a deep learning algorithm to obtain three-dimensional data of the moving object, and obtaining a tailing reconstructed image in the three-dimensional data by using calibration information of a probe;
step D, positioning by using a tailing probe, and superposing a tailing reconstructed image to the area of the tailing probe of the input infrared video image to remove the tailing of the moving object;
and E, denoising the infrared video image with the tail of the moving object removed by adopting a spatial filter and a time domain filter to obtain an output infrared video image.
The method for reducing noise of infrared video based on time-space domain adaptive filtering, wherein the step C further comprises the following steps: dividing a moving image laying grid, and placing a tailing probe in each grid area of a tailing image; the smear probe includes calibration data including an ambiguity of the smear image and a position of a grid of the smear image in the moving image.
The method for reducing noise of the infrared video based on the time-space domain adaptive filtering, wherein the method for judging the fuzzy area of the moving image to obtain the tailing image comprises the following steps: and calculating a fuzzy region of the moving image by using a gray variance algorithm and a Laplacian operator-based fuzzy detection algorithm to obtain a tailing image.
The infrared video noise reduction method based on the time-space domain adaptive filtering comprises a memory module, wherein the memory module comprises a controller, an external image memory and a mask, and the external image memory is used for storing infrared image data filtered by the mask in a first time period.
The filtering processing of the spatial domain filter comprises a spatial distance weight coefficient and a first gray value weight coefficient when the filtering weight value of the spatial neighborhood pixel point is distributed.
The filtering processing of the time domain filter comprises a time distance weight value coefficient and a second gray scale weight coefficient when the filtering weight values of the historical frame and the current frame pixel point are distributed.
The infrared video denoising method based on the time-space domain adaptive filtering comprises the steps of repeatedly and iteratively settling the spatial filter and the time-domain filter to obtain an output infrared video image.
The infrared video noise reduction device based on the time-space domain adaptive filtering is used for removing the tail of an indoor moving object, and comprises a local gray level fluctuation rate image calculation unit, a moving image segmentation unit, a tail reconstruction image calculation unit, a tail removal unit and a noise reduction unit:
the local gray scale fluctuation ratio image calculation unit is used for acquiring pixels of the input infrared video image, calculating local standard images and local mean images of the pixels in the input infrared video image, and calculating local gray scale fluctuation ratio images of the pixels in the input infrared video image according to the local standard images and the local mean images;
the moving image segmentation unit is used for carrying out soft segmentation on the input infrared video image according to the local gray scale fluctuation rate image to obtain a mask of a detail area, and carrying out moving object segmentation on the mask to obtain a moving image of a moving object;
the system comprises a smear reconstruction image calculation unit, a memory module, a motion image processing unit and a motion image processing unit, wherein the smear reconstruction image calculation unit is used for judging a fuzzy area of the motion image to obtain a smear image, the memory module is used for performing deep learning on the motion image, the deep learning algorithm is used for performing super-resolution three-dimensional reconstruction on the motion image to obtain three-dimensional data of a motion object, and the smear reconstruction image in the three-dimensional data is obtained through calibration information of a probe;
the tailing removing unit is used for positioning by using a tailing probe, superposing a tailing reconstructed image to the area of the tailing probe of the input infrared video image and removing the tailing of the moving object;
the noise reduction unit is used for reducing noise of the infrared video image with the tail of the moving object removed by adopting a spatial filter and a time filter to obtain an output infrared video image.
(III) beneficial effects: according to the time-space domain adaptive filtering-based infrared video denoising method and device, three-dimensional stereo data are obtained through three-dimensional reconstruction of a moving object, a tailing area and the three-dimensional stereo data are mapped through a real-time tailing image probe, and a tailing reconstructed image corresponding to the tailing area is obtained, so that the tailing removing calculated amount of the moving object is reduced, and the removing speed is increased. Finally, the invention uses a time domain filtering algorithm and a spatial domain filtering algorithm to reduce noise of the infrared video image after the tail is removed, can effectively solve the blurring phenomenon of the tail edge of the moving object, and improves the definition of the whole output infrared video image.
Drawings
FIG. 1 is a flow chart of an infrared video denoising method based on time-space domain adaptive filtering;
FIG. 2 is a schematic diagram of a memory module according to the present invention;
FIG. 3 is a schematic flow chart of the output infrared video image obtained by denoising the infrared video image with the smear removed by adopting a time domain filtering algorithm and a space domain filtering algorithm;
fig. 4 is a schematic structural diagram of an infrared video noise reduction device based on time-space domain adaptive filtering.
Detailed Description
The present invention will be described in further detail with reference to the preferred embodiments, and more details are set forth in the following description in order to provide a thorough understanding of the present invention, but it will be apparent that the present invention can be embodied in many other forms than described herein, and that those skilled in the art may make similar generalizations and deductions depending on the actual application without departing from the spirit of the present invention, and therefore should not be construed to limit the scope of the present invention in the context of this particular embodiment.
The drawings are schematic representations of embodiments of the invention, it being noted that the drawings are by way of example only and are not drawn to scale and should not be taken as limiting the true scope of the invention.
The infrared video noise reduction method based on the time-space domain adaptive filtering is used for removing the tail of an indoor moving object and is mainly applied to removing the tail caused by relative motion between imaging equipment and the object. As shown in fig. 1, the method comprises the following steps:
Acquiring an input image
Said input image->
Comprising pixel positions and pixel values of the image, calculating the input image +.>
Local standard image of middle pixels +.>
And local mean image->
According to local standard image->
And local mean image->
Obtaining local gray scale fluctuation rate image +.>
，
Wherein->
Is a small constant greater than 0 to avoid anomalies where the dividend is 0.
The local mean value is the mean value of pixel values in a local square window taking a certain pixel point as a center, and each pixel point is provided with a rolling local window (the edge of the image is expanded outwards) corresponding to one local mean value, and the local mean value corresponding to the pixel of the input image forms a local mean value image.
The local standard value of the invention is the standard deviation of the pixel value in a local square window taking a certain pixel point as the center, and the local standard value of the pixel of the input image forms a local standard image.
And 102, performing soft segmentation on the input infrared video image according to the local gray scale fluctuation rate image to obtain a mask of a detail area, and performing moving object segmentation on the mask to obtain a moving image of the moving object.
And step 103, judging the blurring area of the moving image to obtain a tailing image. And performing deep learning on the moving image by using a memory module, performing super-resolution three-dimensional reconstruction on the moving image by using a deep learning algorithm to obtain three-dimensional data of the moving object, and obtaining a trailing reconstructed image in the three-dimensional data by using calibration information of the probe.
And 104, positioning by using a tailing probe, and superposing a tailing reconstructed image to the area of the tailing probe of the input infrared video image to realize tailing removal of the moving object.
And 105, denoising the infrared video image with the tail of the moving object removed by adopting a spatial filter and a time domain filter to obtain an output infrared video image.
The mask of the detail part of the input infrared video image is obtained through the local gray scale fluctuation rate image, the method is mainly used for removing the tail of the moving object in the image, the area where the moving object is can be accurately obtained by using the mask, and the calculated amount of the image is greatly reduced.
The method for judging the blurring area of the moving image to obtain the tailing image comprises the following steps: and calculating a fuzzy region of the moving image by using a gray variance algorithm and a Laplacian operator-based fuzzy detection algorithm to obtain a tailing image.
The memory module, as shown in fig. 2, includes a controller, an external image memory, and a mask, where the external image memory is configured to store mask-filtered infrared image data during a first period of time, such as moving image data of mask-filtered detail regions having a resolution above a threshold value for 7 days at a current time.
The invention is applied to infrared monitoring of indoor moving objects, wherein the indoor moving objects mainly comprise personnel and articles, and are relatively fixed. Therefore, the present invention can perform three-dimensional reconstruction of moving images by machine learning for these scenes. When the motion image is smeared, the method utilizes the calibration data calibrated by the probe to obtain the smear reconstructed image with high resolution and definition corresponding to the smear image from the three-dimensional reconstructed data.
According to the invention, the infrared video image with the smear removed is subjected to noise reduction by adopting a time domain filtering algorithm and a space domain filtering algorithm to obtain an output infrared video image, as shown in fig. 3, the gray value of the current frame is weighted and estimated by using the gray value of a historical frame in the infrared video image sequence by adopting a bilateral filtering algorithm and a space domain filtering algorithm, and meanwhile, the input infrared video image of the current frame is subjected to bilateral filtering noise reduction treatment, so that the blurring phenomenon of the smear edge of a moving object in the infrared video image with the smear removed can be well solved.
Firstly, inputting an infrared video image into each frame after smear is removed, carrying out bilateral filtering, and then estimating the gray value of the current pixel point by utilizing the information of the neighborhood pixel point so as to eliminate the influence of noise and realize the denoising of a spatial filter; and then estimating the gray value of the current pixel point by utilizing the pixel information of the same position corresponding to the historical frame so as to realize the further noise reduction processing of the time domain filter, wherein the processing range of the historical frame is 20 frames. The input infrared video image is subjected to spatial filter noise reduction processing based on bilateral filtering, the processing result is used as a current frame of the time domain filter, and the processed time domain filter is output and is used as a reference frame of subsequent time domain noise reduction processing.
The invention discloses a filtering process of a spatial filter, which is characterized in that two kinds of information of spatial distance and gray scale difference are considered simultaneously when a filtering weight value of a spatial neighborhood pixel point is distributed, the filtering process of the spatial filter comprises a spatial distance weight coefficient and a first gray scale value weight coefficient when the filtering weight value of the spatial neighborhood pixel point is distributed, so that the closer the spatial distance to the pixel point to be estimated is in a filtering window, the higher the weight value is, the smaller the gray scale value difference is, and the calculation formula of the spatial filter is as follows:
The spatial distance weight coefficient and the first gray value weight coefficient are gaussian functions and are defined as follows:
Equation 2 shows that the spatial distance weight coefficient is of a size and pixels in the neighborhood
And pixels to be estimated->
Is inversely proportional to the distance of the pixel in the neighborhood, equation 3 shows that the first gray value weight is of a magnitude>
And pixels to be estimated +.>
Is inversely proportional to the gray level difference of (c). The spatial filter disclosed by the invention not only considers the pixel distance information in the neighborhood, but also considers the gray information, and as the gray phase difference between pixels at the edge part is larger, the corresponding weight is smaller, so that the spatial filter method disclosed by the invention well protects the edge detail information.
The time interval between different frames in the video image sequence is very small, and the gray level change of most pixels between adjacent frames is relatively small, so that the image information has higher relevance in the time domain, the closer to the current frame, the higher the similarity, and the larger the weight allocated in filtering. The filtering processing of the time domain filter comprises a time distance weight coefficient and a second gray scale weight coefficient when the filtering weight values of the historical frame and the current frame pixel point are distributed. According to the bilateral filtering idea, bilateral Gaussian filtering processing is carried out on the time domain, namely, the time-distance relation between a historical frame and a current frame is considered, the gray value relation between the gray value of each pixel of the current frame and the gray value of the corresponding position in the historical frame is considered, and the calculation formula of the time domain filter function is as follows:
In the above
The frame number is the time domain filtered frame number, which indicates that the time distance weight coefficient is inversely proportional to the time distance between the historical frame and the current frame, and the second gray weight coefficient is inversely proportional to the gray difference between the historical frame and the current frame. As the gray value change between the front frame and the rear frame is larger if the motion area exists in the video, the time domain filter coefficient is well corrected by adding the second gray weight coefficient into the filter, and the smaller the frame weight coefficient with larger second gray difference value, the smaller the influence on the output result is, so that the problem of blurring of the motion trailing edge area is solved.
According to the invention, the output infrared video image is obtained by repeatedly and iteratively settling the space domain filter and the time domain filter.
In the spatial filter of the present invention
=6，
=2>
=5，
Experiments prove that an infrared image video sequence firstly enters a time domain filter through a filtering result frame of a spatial filter, a window of the spatial filter is set to 5*5 in order to improve algorithm instantaneity, and the processing frame number of the time domain filter is>
The gaussian filter coefficient in the time domain filter is actually a half-width one-dimensional gaussian kernel, and the weight of the current frame is highest, and the weight corresponding to the frame with the longer time from the current frame is smaller.
The invention provides an infrared video noise reduction device based on time-space domain adaptive filtering, which is shown in fig. 4 and is used for removing the tail of an indoor moving object, and comprises a local gray scale fluctuation ratio image calculation unit, a moving image segmentation unit, a tail reconstruction image calculation unit, a tail removal unit and a noise reduction unit:
the local gray scale fluctuation ratio image calculation unit is used for acquiring pixels of the input infrared video image, calculating local standard images and local mean images of the pixels in the input infrared video image, and calculating local gray scale fluctuation ratio images of the pixels in the input infrared video image according to the local standard images and the local mean images;
the moving image segmentation unit is used for carrying out soft segmentation on the input infrared video image according to the local gray scale fluctuation rate image to obtain a mask of a detail area, and carrying out moving object segmentation on the mask to obtain a moving image of a moving object;
the system comprises a smear reconstruction image calculation unit, a memory module, a motion image processing unit and a motion image processing unit, wherein the smear reconstruction image calculation unit is used for judging a fuzzy area of the motion image to obtain a smear image, the memory module is used for performing deep learning on the motion image, the deep learning algorithm is used for performing super-resolution three-dimensional reconstruction on the motion image to obtain three-dimensional data of a motion object, and the smear reconstruction image in the three-dimensional data is obtained through calibration information of a probe;
the tailing removing unit is used for positioning by using a tailing probe, superposing a tailing reconstructed image to the area of the tailing probe of the input infrared video image and removing the tailing of the moving object;
the noise reduction unit is used for reducing noise of the infrared video image with the tail of the moving object removed by adopting a spatial filter and a time filter to obtain an output infrared video image.
According to the time-space domain adaptive filtering-based infrared video noise reduction method and device, the local gray scale fluctuation rate is low in the background area and high in the texture area, the local gray scale fluctuation rate image with low gray scale fluctuation rate and high gray scale fluctuation rate in the texture area is generated, background noise amplification can be well restrained, the area where a moving object in an input infrared video image is located can be more accurately identified, and the algorithm has better robustness and controllability. According to the invention, the three-dimensional data is obtained through three-dimensional reconstruction of the moving object, the tailing area and the three-dimensional data are mapped through the probe of the real-time tailing image, and the tailing reconstructed image corresponding to the tailing area is obtained, so that the calculated amount of tailing removal of the moving object is reduced, and the removal speed is increased. Finally, the invention uses a time domain filtering algorithm and a spatial domain filtering algorithm to reduce noise of the infrared video image after the tail is removed, can effectively solve the blurring phenomenon of the tail edge of the moving object, and improves the definition of the whole output infrared video image.
The foregoing is a description of a preferred embodiment of the invention to assist those skilled in the art in more fully understanding the invention. However, these examples are merely illustrative, and the present invention is not to be construed as being limited to the descriptions of these examples. It should be understood that, to those skilled in the art to which the present invention pertains, several simple deductions and changes can be made without departing from the inventive concept, and these should be considered as falling within the scope of the present invention.
Claims (8)
1. The infrared video noise reduction method based on the time-space domain adaptive filtering is used for removing the tail of an indoor moving object and is characterized by comprising the following steps of:
step A, obtaining pixels of an input infrared video image, calculating a local standard image and a local mean value image of the pixels in the input infrared video image, and calculating a local gray scale fluctuation rate image of the pixels in the input infrared video image according to the local standard image and the local mean value image;
b, performing soft segmentation on an input infrared video image according to the local gray scale fluctuation rate image to obtain a mask of a detail area, and performing moving object segmentation on the mask to obtain a moving image of a moving object;
step C, judging a blurring region of the moving image to obtain a tailing image, performing deep learning on the moving image by using a memory module, performing super-resolution three-dimensional reconstruction on the moving image by using a deep learning algorithm to obtain three-dimensional data of the moving object, and obtaining a tailing reconstructed image in the three-dimensional data by using calibration information of a probe;
step D, positioning by using a tailing probe, and superposing a tailing reconstructed image to the area of the tailing probe of the input infrared video image to remove the tailing of the moving object;
and E, denoising the infrared video image with the tail of the moving object removed by adopting a spatial filter and a time domain filter to obtain an output infrared video image.
2. The method for noise reduction of an infrared video based on time-space domain adaptive filtering according to claim 1, wherein said step C further comprises: dividing a moving image laying grid, and placing a tailing probe in each grid area of a tailing image; the smear probe includes calibration data including an ambiguity of the smear image and a position of a grid of the smear image in the moving image.
3. The method for reducing noise of infrared video based on time-space domain adaptive filtering according to claim 1, wherein the method for determining a blurred region of the moving image to obtain a tailing image comprises the following steps: and calculating a fuzzy region of the moving image by using a gray variance algorithm and a Laplacian operator-based fuzzy detection algorithm to obtain a tailing image.
4. The method of infrared video noise reduction based on time-space domain adaptive filtering according to claim 2, wherein the memory module comprises a controller, an external image memory and a mask, wherein the external image memory is used for storing the infrared image data filtered by the mask in the first period of time.
5. The method for noise reduction of an infrared video based on time-space domain adaptive filtering according to claim 4, wherein the filtering process of the spatial filter includes a spatial distance weight coefficient and a first gray value weight coefficient when the filtering weight value of the spatial neighborhood pixel is allocated.
6. The method of claim 5, wherein the filtering of the time-domain filter includes a time-distance weight coefficient and a second gray scale weight coefficient when the filtering weights of the pixels of the historical frame and the current frame are allocated.
7. The method for noise reduction of infrared video based on time-space domain adaptive filtering according to claim 6, wherein the spatial filter and the time-domain filter are iteratively settled repeatedly to obtain an output infrared video image.
8. The infrared video noise reduction device based on the time-space domain adaptive filtering is used for removing the tail of an indoor moving object and is characterized by comprising a local gray level fluctuation rate image calculation unit, a moving image segmentation unit, a tail reconstruction image calculation unit, a tail removal unit and a noise reduction unit:
the local gray scale fluctuation ratio image calculation unit is used for acquiring pixels of the input infrared video image, calculating local standard images and local mean images of the pixels in the input infrared video image, and calculating local gray scale fluctuation ratio images of the pixels in the input infrared video image according to the local standard images and the local mean images;
the moving image segmentation unit is used for carrying out soft segmentation on the input infrared video image according to the local gray scale fluctuation rate image to obtain a mask of a detail area, and carrying out moving object segmentation on the mask to obtain a moving image of a moving object;
the system comprises a smear reconstruction image calculation unit, a memory module, a motion image processing unit and a motion image processing unit, wherein the smear reconstruction image calculation unit is used for judging a fuzzy area of the motion image to obtain a smear image, the memory module is used for performing deep learning on the motion image, the deep learning algorithm is used for performing super-resolution three-dimensional reconstruction on the motion image to obtain three-dimensional data of a motion object, and the smear reconstruction image in the three-dimensional data is obtained through calibration information of a probe;
the tailing removing unit is used for positioning by using a tailing probe, superposing a tailing reconstructed image to the area of the tailing probe of the input infrared video image and removing the tailing of the moving object;
the noise reduction unit is used for reducing noise of the infrared video image with the tail of the moving object removed by adopting a spatial filter and a time filter to obtain an output infrared video image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310312205.2A CN116012265B (en) | 2023-03-28 | 2023-03-28 | Infrared video denoising method and device based on time-space domain adaptive filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310312205.2A CN116012265B (en) | 2023-03-28 | 2023-03-28 | Infrared video denoising method and device based on time-space domain adaptive filtering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116012265A CN116012265A (en) | 2023-04-25 |
| CN116012265B true CN116012265B (en) | 2023-05-16 |
Family
ID=86021421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310312205.2A Active CN116012265B (en) | 2023-03-28 | 2023-03-28 | Infrared video denoising method and device based on time-space domain adaptive filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116012265B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116862918B (en) * | 2023-09-05 | 2023-11-24 | 浙江奔月电气科技有限公司 | Real-time detection method for condensation inside ring main unit based on artificial intelligence |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101241592A (en) * | 2007-02-07 | 2008-08-13 | 南京理工大学 | High frame frequency infrared image sequence movement target real time restoration method |
| JP2011082921A (en) * | 2009-10-09 | 2011-04-21 | Fujifilm Corp | Image capturing apparatus and imaging system |
| CN103024248A (en) * | 2013-01-05 | 2013-04-03 | 上海富瀚微电子有限公司 | Motion-adaptive video image denoising method and device |
| CN110944176A (en) * | 2019-12-05 | 2020-03-31 | 浙江大华技术股份有限公司 | Image frame noise reduction method and computer storage medium |
| CN111583151A (en) * | 2020-05-09 | 2020-08-25 | 浙江大华技术股份有限公司 | Video denoising method and device, and computer readable storage medium |
| CN111968057A (en) * | 2020-08-24 | 2020-11-20 | 浙江大华技术股份有限公司 | Image noise reduction method and device, storage medium and electronic device |
| CN114841866A (en) * | 2022-03-25 | 2022-08-02 | 武汉博宇光电系统有限责任公司 | Infrared image time domain filtering method based on displacement detection registration |
-
2023
- 2023-03-28 CN CN202310312205.2A patent/CN116012265B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101241592A (en) * | 2007-02-07 | 2008-08-13 | 南京理工大学 | High frame frequency infrared image sequence movement target real time restoration method |
| JP2011082921A (en) * | 2009-10-09 | 2011-04-21 | Fujifilm Corp | Image capturing apparatus and imaging system |
| CN103024248A (en) * | 2013-01-05 | 2013-04-03 | 上海富瀚微电子有限公司 | Motion-adaptive video image denoising method and device |
| CN110944176A (en) * | 2019-12-05 | 2020-03-31 | 浙江大华技术股份有限公司 | Image frame noise reduction method and computer storage medium |
| CN111583151A (en) * | 2020-05-09 | 2020-08-25 | 浙江大华技术股份有限公司 | Video denoising method and device, and computer readable storage medium |
| CN111968057A (en) * | 2020-08-24 | 2020-11-20 | 浙江大华技术股份有限公司 | Image noise reduction method and device, storage medium and electronic device |
| CN114841866A (en) * | 2022-03-25 | 2022-08-02 | 武汉博宇光电系统有限责任公司 | Infrared image time domain filtering method based on displacement detection registration |
Non-Patent Citations (1)
| Title |
|---|
| 周克虎,雷涛,罗刚."一种基于时域滤波的红外序列图像去噪算法".《应用光学》.2021,(第3期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116012265A (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102326379B (en) | Method for removing blur from image | |
| CN106920220B (en) | The turbulent flow method for blindly restoring image optimized based on dark primary and alternating direction multipliers method | |
| CN106127688B (en) | A kind of super-resolution image reconstruction method and its system | |
| EP0444874A2 (en) | Improved image processing technique | |
| CN106251297A (en) | A kind of estimation based on multiple image fuzzy core the rebuilding blind super-resolution algorithm of improvement | |
| CN112215773B (en) | Local motion deblurring method and device based on visual saliency and storage medium | |
| WO2017100971A1 (en) | Deblurring method and device for out-of-focus blurred image | |
| CN106097256B (en) | A kind of video image fuzziness detection method based on Image Blind deblurring | |
| CN105913392A (en) | Degraded image overall quality improving method in complex environment | |
| CN116012265B (en) | Infrared video denoising method and device based on time-space domain adaptive filtering | |
| CN110796616A (en) | Fractional order differential operator based L0Norm constraint and adaptive weighted gradient turbulence degradation image recovery method | |
| CN111899200B (en) | Infrared image enhancement method based on 3D filtering | |
| CN112184567A (en) | Multi-channel blind identification adaptive optical image restoration method based on alternate minimization | |
| CN106023097A (en) | Iterative-method-based flow field image preprocessing algorithm | |
| Li et al. | Infrared small target detection based on 1-D difference of guided filtering | |
| CN110009575B (en) | Infrared image stripe noise suppression method based on sparse representation | |
| US20240005454A1 (en) | Method of performing l0 smoothing on the basis of deep gradient prior information to improve sharpness | |
| US10181181B2 (en) | Denoising with three dimensional fourier transform for three dimensional images, including image sequences | |
| Shankar et al. | Object oriented fuzzy filter for noise reduction of Pgm images | |
| Zeng et al. | Noise suppression by discontinuity indicator controlled non-local means method | |
| CN113822823A (en) | Point neighbor restoration method and system for aerodynamic optical effect image space-variant fuzzy core | |
| CN113269687A (en) | Non-local average infrared image denoising method based on parameter optimization | |
| CN115861134B (en) | Star map processing method under space-based background | |
| Yang et al. | Non-blind image blur removal method based on a Bayesian hierarchical model with hyperparameter priors | |
| CN116091360B (en) | Non-uniformity noise elimination method and device based on surface fitting |
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 |