CN113538255A - Motion fusion noise reduction method and device and computer readable storage medium - Google Patents
Motion fusion noise reduction method and device and computer readable storage medium Download PDFInfo
- Publication number
- CN113538255A CN113538255A CN202110604899.8A CN202110604899A CN113538255A CN 113538255 A CN113538255 A CN 113538255A CN 202110604899 A CN202110604899 A CN 202110604899A CN 113538255 A CN113538255 A CN 113538255A
- Authority
- CN
- China
- Prior art keywords
- noise reduction
- image
- visible light
- intensity
- current
- 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 429
- 230000004927 fusion Effects 0.000 title claims abstract description 111
- 238000000034 method Methods 0.000 title claims abstract description 67
- 230000003068 static effect Effects 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims description 46
- 230000003287 optical effect Effects 0.000 claims description 13
- 230000002123 temporal effect Effects 0.000 claims description 11
- 239000002131 composite material Substances 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 description 9
- 230000000007 visual effect Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 239000000047 product Substances 0.000 description 4
- 239000013589 supplement Substances 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
Abstract
The invention discloses a motion fusion noise reduction method, equipment and a computer readable storage medium, wherein the method comprises the following steps: acquiring a current visible light image and a current infrared light image of a target; performing time domain noise reduction on the current visible light image to obtain a current visible light noise reduction image, wherein the visible light time domain noise reduction intensity of the motion region of the current visible light image is smaller than the visible light time domain noise reduction intensity of the static region of the current visible light image; and fusing the motion area image of the current infrared light image and the motion area image of the current visible light noise reduction image to obtain a current visible light fused image corresponding to the motion area. By the mode, the method and the device can make up for the defect that the image details cannot be supplemented by effectively utilizing the time domain information in the motion area of the current visible light image, so that the detail information of the motion area of the current visible light image is better.
Description
Technical Field
The present invention relates to the field of image processing, and in particular, to a motion fusion noise reduction method, device, and computer-readable storage medium.
Background
In a low-light scene, a visible light image presented by a common camera is often affected by noise, and a noise reduction technology is needed to reduce the noise of the image so as to improve the visual effect. In recent years, the technology for guiding the visible light image to reduce noise by using the infrared light image is more and more widely applied, but the guiding and the noise reduction of the visible light image by using the infrared light image are difficult, improper guiding cannot improve the visual effect, and some abnormalities can be caused. If when the time domain noise reduction is performed on the image, the risk of the smearing effect is reduced by reducing the noise reduction intensity of the motion region, but the noise of the motion region is more obvious than that of the static region, so that the noise suppression performance of the motion region is poorer as a noise reduction result under the action of the time domain noise reduction.
Disclosure of Invention
The invention mainly solves the technical problem of providing a motion fusion noise reduction method, a device and a computer readable storage medium, which can make up the defect that the motion area of the current visible light image cannot effectively utilize time domain information to supplement image details, so that the detail information of the motion area of the current visible light image is better.
In order to solve the technical problems, the invention adopts a technical scheme that: provided is a motion fusion noise reduction method, comprising: acquiring a current visible light image and a current infrared light image of a target; performing time domain noise reduction on the current visible light image to obtain a current visible light noise reduction image, wherein the visible light time domain noise reduction intensity of the motion region of the current visible light image is smaller than the visible light time domain noise reduction intensity of the static region of the current visible light image; and fusing the motion area image of the current infrared light image and the motion area image of the current visible light noise reduction image to obtain a current visible light fused image corresponding to the motion area.
Wherein, after carrying out time domain noise reduction on the current visible light image, the method comprises the following steps: and fusing the static area image of the current infrared light image and the static area image of the current visible light noise reduction image to obtain a current visible light fused image corresponding to the static area.
The method for fusing the moving area image of the current infrared light image with the moving area image of the current visible light noise reduction image to obtain the current visible light fused image of the corresponding moving area comprises the following steps: acquiring a first guiding intensity and a second guiding intensity of a current infrared image, wherein the first guiding intensity is greater than the second guiding intensity; the first guiding intensity is obtained from a first area image of the current infrared light image, the second guiding intensity is obtained from a second area image of the current infrared light image, and the signal-to-noise ratio of the first area image of the current infrared light image is higher than that of the second area image of the current infrared light image; carrying out weighted fusion on the motion region image of the current infrared light image and the motion image of the current visible light noise reduction image; carrying out weighted fusion on the static area image of the current infrared light image and the static area image of the current visible light noise reduction image; the weight of the moving area image of the current infrared image is a first fusion intensity, the weight of the static area image of the current infrared image is a second fusion intensity, the first fusion intensity and the second fusion intensity are obtained by calculation respectively through the first guide intensity and the second guide intensity, and the first fusion intensity is larger than the second fusion intensity.
The method comprises the following steps of fusing a motion area image of a current infrared light image with a motion area image of a current visible light noise reduction image, wherein the fusion comprises the following steps: and carrying out noise reduction processing on the current infrared light image, wherein the noise reduction processing comprises time domain noise reduction and/or spatial domain noise reduction.
Wherein, before performing time domain noise reduction on the current visible light image, the method comprises the following steps: and performing spatial domain noise reduction on the current visible light image.
Wherein, after carrying out time domain noise reduction on the current visible light image, the method comprises the following steps: and performing spatial domain noise reduction on the current visible light image.
Wherein the time-domain denoising of the current visible light image comprises: acquiring a plurality of historical visible light noise reduction images of a target, wherein the historical visible light noise reduction images are obtained by performing noise reduction processing on the historical visible light images; acquiring comprehensive visible light time domain noise reduction intensity by using the plurality of historical visible light noise reduction images and the current visible light image; and performing time domain noise reduction on the current visible light image by combining the comprehensive visible light time domain noise reduction intensity and the time domain noise reduction algorithm.
Wherein the noise reduction processing of the historical visible light image comprises: sequentially carrying out time domain noise reduction processing and spatial domain noise reduction processing on the historical visible light image; or sequentially performing spatial domain noise reduction processing and time domain noise reduction processing on the historical visible light image.
The method for acquiring the comprehensive visible light time domain noise reduction intensity by using the plurality of historical visible light noise reduction images and the current visible light image comprises the following steps:
acquiring infrared light time domain noise reduction intensity of an infrared light image and initial visible light time domain noise reduction intensity of a current visible light image, wherein the infrared light time domain noise reduction intensity of the infrared light image comprises a first infrared light time domain noise reduction intensity and a second infrared light time domain noise reduction intensity, the first infrared light time domain noise reduction intensity is the noise reduction intensity for performing time domain noise reduction processing on a first area image of the infrared light image, the second infrared light time domain noise reduction intensity is the noise reduction intensity for performing time domain noise reduction processing on a second area image of the infrared light image, and the signal-to-noise ratio of the first area image of the infrared light image is higher than the signal-to-noise ratio of the second area image of the infrared light image; the initial visible light time domain noise reduction intensity of the current visible light image comprises a first initial visible light time domain noise reduction intensity and a second initial visible light time domain noise reduction intensity, the first initial visible light time domain noise reduction intensity is the time domain noise reduction intensity of a first area image corresponding to the visible light image, and the second initial visible light time domain noise reduction intensity is the time domain noise reduction intensity of a second area image corresponding to the visible light image; carrying out weighted fusion on the first infrared light time domain noise reduction intensity and the first initial visible light time domain noise reduction intensity; carrying out weighted fusion on the second infrared light time domain noise reduction intensity and the second initial visible light time domain noise reduction intensity to obtain comprehensive visible light time domain noise reduction intensity; the weight of the first infrared light time domain noise reduction intensity is a first guide intensity, and the weight of the second infrared light time domain noise reduction intensity is a second guide intensity; wherein the first guiding intensity is obtained from a first area image of the current infrared light image, and the second guiding intensity is obtained from a second area image of the current infrared light image.
Acquiring infrared light airspace noise reduction intensity of an infrared light image and initial visible light airspace noise reduction intensity of a current visible light image, wherein the infrared light airspace noise reduction intensity of the infrared light image comprises first infrared light airspace noise reduction intensity and second infrared light airspace noise reduction intensity, the first infrared light airspace noise reduction intensity is the noise reduction intensity for performing airspace noise reduction processing on a first area image of the infrared light image, the second visible light airspace noise reduction intensity is the noise reduction intensity for performing airspace noise reduction processing on a second area image of the infrared light image, and the signal-to-noise ratio of the first area image of the infrared light image is higher than the signal-to-noise ratio of the second area image of the infrared light image; the initial visible light spatial noise reduction intensity of the current visible light image comprises a first initial visible light spatial noise reduction intensity and a second initial visible light spatial noise reduction intensity, wherein the first initial visible light spatial noise reduction intensity is the spatial noise reduction intensity of a first region image corresponding to the visible light image, and the second initial visible light spatial noise reduction intensity is the spatial noise reduction intensity of a second region image corresponding to the visible light image; carrying out weighted fusion on the first infrared light spatial noise reduction intensity and the first initial visible light spatial noise reduction intensity; weighting and fusing the second infrared light airspace noise reduction intensity and the second initial visible light airspace noise reduction intensity to obtain comprehensive visible light airspace noise reduction intensity; the weight of the first infrared airspace noise reduction intensity is a first guide intensity, and the weight of the second infrared airspace noise reduction intensity is a second guide intensity; the first guiding intensity is obtained from a first area image of the current infrared light image, and the second guiding intensity is obtained from a second area image of the current infrared light image; and performing spatial domain noise reduction on the current visible light image by combining the comprehensive visible light spatial domain noise reduction intensity and a spatial domain noise reduction algorithm.
The method includes the following steps that fusion is carried out on a current infrared light image of a motion area and a current visible light noise reduction image of the motion area, and after the current visible light fusion image of the corresponding motion area is obtained, the method further includes the following steps: and performing spatial domain noise reduction on the visible light noise reduction image.
In order to solve the technical problem, the invention adopts another technical scheme that: a motion fusion noise reduction device is provided, which comprises a processor for executing instructions to realize the motion fusion noise reduction method.
In order to solve the technical problem, the invention adopts another technical scheme that: a computer readable storage medium is provided for storing instructions/program data that can be executed to implement the motion fusion noise reduction method described above.
The invention has the beneficial effects that: different from the situation of the prior art, the method provided by the invention has the advantages that after the current visible light image is subjected to spatial domain noise reduction, the motion area image of the current visible light noise reduction image is fused with the corresponding motion area image of the current infrared light image, so that the defect that the motion area of the current visible light image cannot effectively utilize time domain information to supplement image details is overcome, the detailed information of the motion area of the current visible light image is better, and the visual impression is better.
Drawings
FIG. 1 is a schematic flow chart of a motion fusion noise reduction method according to an embodiment of the present disclosure;
FIG. 2 is a schematic flow chart of another motion fusion noise reduction method according to an embodiment of the present disclosure;
FIG. 3 is a schematic flow chart of a method for fusing a current infrared light image and a current visible light noise reduction image according to an embodiment of the present disclosure;
fig. 4 is a schematic flowchart of noise reduction processing performed on a current infrared light image in the embodiment of the present application;
fig. 5 is a schematic flowchart of noise reduction processing performed on a current visible light image in the embodiment of the present application;
FIG. 6 is a schematic structural diagram of a motion fusion noise reduction apparatus according to an embodiment of the present disclosure;
FIG. 7 is a schematic structural diagram of a motion fusion noise reduction device in an embodiment of the present application;
fig. 8 is a schematic structural diagram of a computer-readable storage medium in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and effects of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples.
Referring to fig. 1, fig. 1 is a schematic flow chart of a motion fusion noise reduction method according to an embodiment of the present disclosure. It should be noted that, if the result is substantially the same, the flow sequence shown in fig. 1 is not limited in this embodiment. As shown in fig. 1, the present embodiment includes:
s110: and acquiring a current visible light image and a current infrared light image of the target.
In this embodiment, the device that captures the visible light image and the infrared light image simultaneously may be used to obtain the visible light image and the infrared light image of the target simultaneously, such as a thermal infrared imager. It is also possible to use different devices to acquire the visible light image and the infrared light image of the object separately, such as a visible light camera and an infrared camera. When different devices are used for acquiring the visible light image and the infrared light image, the two devices can be placed at the same position, and the optical axes of the lenses are in the same direction and parallel, so that the visible light image and the infrared light image at the same angle can be acquired. The two devices can also be placed at different positions to acquire visible light images and infrared light images at different angles. The embodiment does not limit the device used and the image acquisition angle. And taking the visible light image of the current frame subjected to processing as a current visible light image, and taking an infrared light image corresponding to the current visible light as a current infrared light image.
S130: and performing time domain noise reduction on the current visible light image to obtain the current visible light noise reduction image.
The image may be divided into a motion area and a still area according to different motion information, and in an embodiment of the present application, the motion area and the still area may be divided according to a motion information detection method, or may also be considered as defining the motion area and the still area, which is not limited herein. And setting different time domain noise reduction intensities in the motion area and the static area of the current visible light image, and performing time domain noise reduction on the current visible light image to obtain the current visible light noise reduction image. And the visible light time domain noise reduction intensity of the motion region of the current visible light image is smaller than the visible light time domain noise reduction intensity of the static region of the current visible light image.
S150: and fusing the motion area of the current infrared light image with the motion area of the current visible light noise reduction image to obtain a current visible light fused image corresponding to the motion area.
And introducing the information of the motion area of the current infrared light image into the motion area of the current visible light noise reduction image, and fusing the motion area of the current infrared light image and the motion area of the current visible light noise reduction image to obtain the current visible light fusion image corresponding to the motion area.
In the embodiment, after the spatial domain noise reduction is performed on the current visible light image, the motion region image of the current visible light noise reduction image is fused with the corresponding motion region image of the current infrared light image, so that the defect that the motion region of the current visible light image cannot supplement image details by effectively utilizing time domain information is overcome, and the detailed information of the motion region of the current visible light image is better, and the visual impression is better.
Referring to fig. 2, fig. 2 is a schematic flow chart of another motion fusion noise reduction method according to an embodiment of the present application. It should be noted that, if the result is substantially the same, the flow sequence shown in fig. 2 is not limited in this embodiment. As shown in fig. 2, the present embodiment includes:
s210: and acquiring a current visible light image and a current infrared light image of the target.
S230: and performing time domain noise reduction on the current visible light image to obtain the current visible light noise reduction image.
S250: fusing the current infrared light image of the motion area with the current visible light noise reduction image of the motion area to obtain a current visible light fused image of the corresponding motion area; and fusing the current infrared light image of the static area and the current visible light noise reduction image of the static area to obtain a current visible light fusion image corresponding to the static area.
And introducing the information of the motion area of the current infrared light image into the motion area of the current visible light noise reduction image, and simultaneously introducing the information of the static area of the current infrared light image into the static area of the current visible light noise reduction image.
When the current infrared light image and the current visible light noise reduction image are fused, the original current visible light image without noise reduction processing or the current infrared light image with noise reduction processing may be used, and the noise reduction method may be spatial domain noise reduction, temporal domain noise reduction, a combination of spatial domain noise reduction and temporal domain noise reduction, or other noise reduction methods, which are not limited herein.
S270: and performing spatial domain noise reduction on the current visible light fusion image.
After the whole current infrared light image and the current visible light image are fused, the current visible light fused image is subjected to sequential spatial domain noise reduction treatment, and the specific noise reduction method is not limited.
Referring to fig. 3, fig. 3 is a schematic flow chart illustrating a method for fusing a current infrared light image and a current visible light noise reduction image according to an embodiment of the present disclosure. It should be noted that, if the result is substantially the same, the flow sequence shown in fig. 3 is not limited in this embodiment. As shown in fig. 3, the present embodiment includes:
s310: and acquiring a first guiding intensity and a second guiding intensity of the current infrared light image.
The important index for measuring the image quality in the image signal to noise ratio is the ratio of the size of the video signal to the size of the noise signal, and the image area with high signal to noise ratio has low noise and the image area with low signal to noise ratio has high noise. Different areas of the current infrared light image have different signal-to-noise ratios, different guiding intensities are set according to the difference of the signal-to-noise ratios, larger guiding intensity is set in an area with a high signal-to-noise ratio, and smaller guiding intensity is set in an area with a low signal-to-noise ratio. Alternatively, the same guiding intensity may be set for the whole infrared light, or guiding intensities of different gradient sizes may be set according to the signal-to-noise ratio gradient.
According to different signal-to-noise ratios, the current infrared light image is divided into a first area and a second area, and the signal-to-noise ratio of the current infrared light image in the first area is higher than that of the current infrared light image in the second area. The first guide strength of the first area image is increased as a whole, and the second guide strength of the second area image is decreased as a whole. In the embodiment of the present application, the first region and the second region may be determined according to luminance information, chromaticity information, and the like of the current infrared light image.
In one embodiment, the first region and the second region are determined according to brightness information of the current infrared light image. The first threshold th1, the second threshold th2, and the third threshold th3 are preset, and the area in the current infrared light image with the brightness value greater than th1 is used as a potential overexposure area, the area in the current infrared light image with the brightness value less than th2 is used as a potential information loss area, and the other areas are used as normal areas. And calculating variance information of the overexposure area and the information loss area, wherein the variance information can be the variance information of each pixel point or the variance information of a plurality of pixel areas, if the variance information is higher than a third threshold th3, the variance information indicates that the brightness of the image in the area is too high or too low, the image is caused by the reflection characteristic of the material, and the image itself is normal, otherwise, the area is an information abnormal area. The normal area is set as a first area, the information abnormal area is set as a second area, a first guidance intensity is set in the first area, and a second guidance intensity is set in the second area, and the guidance intensity may be set or calculated based on other information, which is not limited herein.
The first threshold th1, the second threshold th2, and the third threshold th3 are preset thresholds, and default values are determined according to actual situations, in an embodiment, the first threshold th1 is 240, the second threshold th2 is 10, the third threshold th3 is 5, the calculated guidance strength is gui _ ratio, and a value range of the guidance strength is greater than or equal to 0 and less than or equal to 1.
In another embodiment, the guiding intensity may be set according to the brightness information and the chromaticity information of the current infrared light image subjected to the noise reduction processing; the guiding intensity can be calculated by combining the current infrared light image without noise reduction processing and the current infrared light image after noise reduction processing; the guiding intensity may also be calculated jointly in connection with the current visible light image and the current infrared light image.
S330: and performing weighted fusion on the current infrared light image and the current visible light noise reduction image by using the first guide intensity and the second guide intensity.
In this embodiment, the current infrared light image weighted and fused with the current visible light noise reduction image may be the initial current infrared light image or an image subjected to noise reduction processing. The current infrared light image may be denoised by the above denoising method, which is not described herein again.
According to the motion information of the current visible light image and the current infrared light image, different initial fusion intensities are set for the areas with different motion information, and the initial fusion intensity of the motion area is larger than that of the static area. The range of the initial fusion strength is 0 or more and 1 or less. In one embodiment, a larger initial first fusion strength is set in the motion region, and the value of the initial first fusion strength is biased to 1, and a smaller initial second fusion strength is set in the static region, and the value of the initial second fusion strength is biased to 0.
The initial fusion strength is adjusted according to the guiding strength. Firstly, multiplying a preset fourth threshold th4 by the guidance strength to obtain an adjusted guidance strength, wherein the adjusted guidance strength is gui _ th, the guidance strength greater than 1 in the adjusted guidance strength is reset to 1, the guidance strength less than 0 in the adjusted guidance strength is reset to 0, and the value range of the guidance strength is maintained to be greater than or equal to 0 and less than or equal to 1. And multiplying the guiding intensity of the corresponding pixel point by the initial fusion intensity to obtain the adjusted fusion intensity, which is fusion _ ratio. Wherein the fourth threshold th4 is a predetermined threshold, and optionally the default value of the fourth threshold th4 is 1. After adjustment, the fusion intensity of the motion area is the first fusion intensity, the fusion intensity of the static area is the second fusion intensity, and the first fusion intensity is larger than the second fusion intensity.
The current infrared light image and the current visible light noise reduction image are fused by using the adjusted fusion intensity, and various fusion modes can be used. In one embodiment, the motion region image of the current infrared light image and the motion region image of the current visible light noise reduction image are subjected to weighted fusion, and the weight of the motion region image of the current infrared light image is a first fusion intensity; and performing weighted fusion on the static area image of the current infrared light image and the static area image of the current visible light noise reduction image, wherein the weight of the static area image of the current infrared light image is the second fusion intensity, and obtaining the current visible light fusion image. The current visible light noise reduction image is divided into a current visible light brightness channel noise reduction image and a current visible light chromaticity channel noise reduction image. The current visible light luminance channel noise reduction image and the current visible light chrominance channel noise reduction image are respectively fused with the current infrared light image in the same way. The specific fusion formula is as follows:
vis_0＝(1-fusion_ratio)×vis+fusion_ratio×nir
wherein, vis is a current visible light brightness channel noise reduction image, nir is a current infrared image, and vis _0 is a current visible light fusion image.
In the embodiment, different guiding intensities are set according to different signal-to-noise ratios of regions of the current infrared image, different fusion intensities are set according to different motion information of the current visible image, the current infrared image, a motion region and a static region, the fusion intensities are adjusted by using the guiding intensities, more information of the current infrared image is fused in the motion region of the current visible noise reduction image, and less information of the current infrared image is fused in the static region, so that more detailed information can be supplemented in the motion region, the visual impression is better, the previous information can be fully utilized by reducing the fusion intensity in the static region, and the fidelity of brightness information is guaranteed.
In the embodiment of the application, the current infrared light image to be subjected to image fusion may be an image subjected to noise reduction, and the noise reduction method includes spatial noise reduction, temporal noise reduction, a combination of spatial noise reduction and temporal noise reduction, or other noise reduction methods. Referring to fig. 4, fig. 4 is a schematic flow chart illustrating a noise reduction process performed on a current infrared image according to an embodiment of the present disclosure. It should be noted that, if the result is substantially the same, the flow sequence shown in fig. 4 is not limited in this embodiment. As shown in fig. 4, the present embodiment includes:
s410: and carrying out motion detection on the current infrared image to obtain the motion information of the current infrared image.
The current infrared light image motion information can be obtained by performing motion detection on the current infrared light image by using various motion detection methods, such as a frame difference method, an optical flow method, a background difference method, and the like. In one embodiment, a frame difference method is used to perform motion detection on a current infrared light image, and two frames of infrared light images before and after the current infrared light image are compared, wherein if the difference between pixels of two frames of images before and after an area is larger, the area is more biased to a motion area, and if the difference between pixels of two frames of images before and after an area is smaller, the area is more biased to a static area. In another embodiment, the current infrared light image is subjected to motion detection using an optical flow method, and if the optical flow value of an area is large, the area is more likely to be a motion area, and if the optical flow value of an area is small, the area is more likely to be a still area.
S430: and performing spatial domain noise reduction on the current infrared light image.
The spatial domain noise reduction intensity of the current infrared light image can be obtained according to various image information of the current infrared light image. In one embodiment, the current infrared light image spatial domain noise reduction intensity can be determined by using image information such as variance, edge and the like, and when the variance information of a region image is larger, the noise of the region image is larger, so that the larger current infrared light image spatial domain noise reduction intensity is set in the region to reduce the noise of the region in the subsequent spatial domain noise reduction; and when the edge information of an area image is stronger, setting smaller current infrared light image spatial noise reduction intensity in the area so as to keep the large edge information of the area in the subsequent spatial noise reduction.
And adjusting the acquired airspace noise reduction intensity of the current infrared image according to the motion information of the current infrared image. And differentiating the spatial domain noise reduction intensity of the current infrared image according to different motion information, if the current infrared image comprises an area A and an area B, deciding that the spatial domain noise reduction intensity of the front infrared image of the area A is the same as that of the front infrared image of the area B by using variance information, wherein the area A is a motion area, and the area B is a static area, properly increasing the spatial domain noise reduction intensity of the front infrared image of the area A, and reducing the spatial domain noise reduction intensity of the front infrared image of the area B to obtain the adjusted spatial domain noise reduction intensity of the current infrared image, i.e. nirspa _ ratio.
And according to the adjusted airspace noise reduction intensity of the current infrared image, performing noise reduction on the current infrared image by using an airspace noise reduction algorithm, wherein the specific airspace noise reduction algorithm is not limited herein.
S450: and performing time domain noise reduction on the current infrared light image.
And acquiring a plurality of historical infrared light noise reduction images of the target, wherein the historical infrared light noise reduction images are obtained by performing noise reduction processing on the historical infrared light images. In one embodiment, time-domain denoised infrared light denoising images of a plurality of forward frame images are obtained, and the time-domain denoising strength of the current infrared light image is determined by combining the current infrared light image and a plurality of historical infrared light denoising images.
And adjusting the acquired time domain noise reduction intensity of the current infrared image according to the motion information of the current infrared image. According to the difference of the motion information, the time domain noise reduction intensity of the current infrared light image in the motion area is reduced, even the time domain noise reduction intensity of the current infrared light image in the motion area can be set to zero, and the time domain noise reduction intensity of the current infrared light image in the static area is increased. In one embodiment, for a motion region, the inverse of the motion information of the current infrared light image is taken as the temporal noise reduction intensity of the current infrared light image. The adjusted time-domain noise reduction intensity of the current infrared light image is nittimR _ ratio.
And according to the adjusted time domain noise reduction intensity of the current infrared image, performing noise reduction on the current infrared image by using a time domain noise reduction algorithm, wherein the specific time domain noise reduction algorithm is not limited herein.
However, only one of the steps S410 and S430 may be performed, or both of the steps S410 and S430 may be continuously performed. The specific sequence of S410 and S430 is not limited, if the spatial domain noise reduction of S410 is performed first, the time domain noise reduction is performed on the current infrared light spatial domain noise reduction image after the spatial domain noise reduction in S430; if the time-domain noise reduction of S430 is performed first, the current infrared light time-domain noise reduction image after the time-domain noise reduction is performed in S410.
S470: and optimizing the current infrared image to obtain the noise-reduced current infrared image.
And performing post-processing operation on the current infrared light image, and substantially performing primary airspace noise reduction on the current infrared light image to obtain the current infrared light noise reduction image. The specific noise reduction method is the same as S410, and is not described herein again.
In the embodiment, the current infrared light image is subjected to space-domain and time-domain noise reduction, and the output part of the time-domain noise reduction is also used as the historical infrared light image input during the time-domain noise reduction of the next frame, so that the detail information after the time-domain noise reduction can be ensured not to be lost during the time-domain noise reduction of the next frame of image. And then, performing spatial domain noise reduction again on the basis of the noise reduction result, and reducing the requirements on the spatial domain noise reduction intensity and the time domain noise reduction intensity which are performed before while optimizing the final infrared light image noise reduction effect. By the method, the visual effect and the impression of the finally presented current infrared light noise reduction image are better.
In the embodiment of the present application, the current visible light image to be subjected to image fusion may be an image subjected to only temporal noise reduction processing, or may be an image subjected to spatial noise reduction and temporal noise reduction processing in combination. Referring to fig. 5, fig. 5 is a schematic flowchart illustrating a process of performing noise reduction on a current visible light image according to an embodiment of the present disclosure. It should be noted that, if the result is substantially the same, the flow sequence shown in fig. 5 is not limited in this embodiment. As shown in fig. 5, the present embodiment includes:
s510: and carrying out motion detection on the current visible light image to obtain the motion information of the current visible light image.
The current visible light image may be divided into a current visible light luminance channel image and a current visible light chromaticity channel image, and the current visible light luminance channel image and the current visible light chromaticity channel image are respectively subjected to motion detection, and may be detected by using a motion detection method such as a frame difference method, an optical flow method, a background difference method, and the like, so as to obtain motion information of the current visible light luminance channel image and the current visible light chromaticity channel image. The specific acquiring method is the same as that for acquiring the motion information of the current infrared light image in S410, and is not described herein again.
S530: and performing spatial domain noise reduction on the current visible light image.
The current visible light image airspace noise reduction intensity can be obtained according to various image information of the current visible light image, and then the obtained current visible light airspace noise reduction intensity is adjusted according to the motion information of the current visible light image. The specific obtaining method is the same as the method for obtaining the adjusted current infrared light image spatial domain noise reduction intensity in S430, and details are not repeated here.
And adjusting the noise reduction intensity of the current visible airspace for the second time by using the guide intensity. The fifth threshold th5 and the sixth threshold th6 are preset, the product of the first guiding strength or the fifth threshold th5 and the first guiding strength is used as a first weight, the product of the second guiding strength or the sixth threshold th6 and the second guiding strength is used as a second weight, the value larger than 1 in the first weight and the second weight is reset to 1, and the value smaller than 0 is reset to 0, so that the initial visible light spatial noise reduction strength is obtained. Wherein, the obtained first weight is nor _ th, and the obtained second weight is unnor _ th.
Taking the current infrared light airspace noise reduction intensity of a first area of the current infrared light image as a first current infrared light airspace noise reduction intensity, and taking the initial visible light airspace noise reduction intensity of a corresponding area of the current visible light image as a first initial visible light airspace noise reduction intensity; and taking the current infrared light airspace noise reduction intensity of a second region of the current infrared light image as a second infrared light airspace noise reduction intensity, and taking the initial visible light airspace noise reduction intensity of a corresponding region of the current visible light image as a second initial visible light airspace noise reduction intensity. Weighting and fusing the first infrared light airspace noise reduction intensity and the first initial visible light airspace noise reduction intensity, wherein the fusion weight of the first infrared light airspace noise reduction intensity is a first weight; and performing weighted fusion on the second infrared light airspace noise reduction intensity and the second initial visible light airspace noise reduction intensity, wherein the fusion weight of the second infrared light airspace noise reduction intensity is the second weight, and the comprehensive visible light airspace noise reduction intensity is obtained.
The current visible light noise reduction image is divided into a current visible light brightness channel noise reduction image and a current visible light chromaticity channel noise reduction image. The spatial domain noise reduction is performed on the current visible light luminance channel image and the current visible light chromaticity channel image of the current visible light image, and the following embodiment takes the spatial domain noise reduction performed on the current visible light luminance channel image as an example. The specific fusion formula is as follows:
wherein visspa _ ratio is the spatial noise reduction intensity of the current visible light brightness channel noise reduction image, nirspa _ ratio is the spatial noise reduction intensity of the current infrared light image, and visspa _ ratio _0 is the spatial noise reduction intensity of the current visible light image.
And according to the adjusted spatial domain noise reduction intensity of the current visible light image, performing noise reduction on the current visible light image by using a spatial domain noise reduction algorithm, wherein the specific spatial domain noise reduction algorithm is not limited herein.
S550: and performing time domain noise reduction on the current visible light image.
And acquiring a plurality of historical visible light noise reduction images of the target, wherein the historical visible light noise reduction images are obtained by performing noise reduction processing on the historical visible light images. In one embodiment, a visible light noise reduction image of a plurality of forward frame images, which is subjected to spatial domain noise reduction processing and time domain noise reduction processing in sequence, is obtained, and the initial visible light image time domain noise reduction intensity is determined by combining the current visible light image and a plurality of historical visible light noise reduction images. In another embodiment, a visible light noise reduction image of a plurality of forward frame images, which is subjected to time domain noise reduction processing and spatial domain noise reduction processing in sequence, is obtained, and the initial visible light image time domain noise reduction intensity is decided by combining the current visible light image and a plurality of historical visible light noise reduction images.
The time domain noise reduction intensity of the current visible light image can be obtained according to various image information of the current visible light image, and then the obtained time domain noise reduction intensity of the current visible light image is adjusted according to the motion information of the current visible light image. The specific obtaining method is the same as the time-domain noise reduction intensity of the current infrared image obtained and adjusted in S450, and is not described herein again.
And adjusting the noise reduction intensity of the current visible time domain for the second time by using the guide intensity. Setting a seventh threshold th7 and an eighth threshold th8 in advance, taking the product of the first guiding strength or the seventh threshold th7 and the first guiding strength as a third weight, taking the product of the second guiding strength or the eighth threshold th8 and the second guiding strength as a fourth weight, resetting the value greater than 1 in the third weight and the fourth weight as 1, and resetting the value less than 0 as 0 to obtain the initial visible light time domain noise reduction strength. Wherein, the obtained third weight is nor _ th1, and the obtained fourth weight is unnor _ th 1.
Taking the current infrared light time domain noise reduction intensity of a first area of the current infrared light image as a first current infrared light time domain noise reduction intensity, and taking the initial visible light time domain noise reduction intensity of a corresponding area of the current visible light image as a first initial visible light time domain noise reduction intensity; and taking the current infrared light time domain noise reduction intensity of the second region of the current infrared light image as a second infrared light time domain noise reduction intensity, and taking the initial visible light time domain noise reduction intensity of the corresponding region of the current visible light image as a second initial visible light time domain noise reduction intensity. Performing weighted fusion on the first infrared optical time domain noise reduction intensity and the first initial visible light time domain noise reduction intensity, wherein the fusion weight of the first infrared optical time domain noise reduction intensity is a first weight; and performing weighted fusion on the second infrared optical time domain noise reduction intensity and the second initial visible light time domain noise reduction intensity, wherein the fusion weight of the second infrared optical time domain noise reduction intensity is the second weight, and the comprehensive visible light time domain noise reduction intensity is obtained.
The current visible light noise reduction image is divided into a current visible light brightness channel noise reduction image and a current visible light chromaticity channel noise reduction image. The time domain noise reduction is performed on the current visible light luminance channel image and the current visible light chromaticity channel image of the current visible light image, and the time domain noise reduction is performed on the current visible light luminance channel image in the following embodiment as an example. The specific fusion formula is as follows:
wherein, visspa _ ratio _1 is the temporal denoising intensity of the current visible light brightness channel denoising image, nirspa _ ratio _1 is the temporal denoising intensity of the current infrared light image, and visspa _ ratio _1 is the temporal denoising intensity of the current visible light image.
And according to the adjusted time domain noise reduction intensity of the current visible light image, performing noise reduction on the current visible light image by using a time domain noise reduction algorithm, wherein the specific time domain noise reduction algorithm is not limited herein.
In this embodiment, by setting different guide intensities according to different signal-to-noise ratios of respective regions of the current infrared light image, the quality of the infrared light image can be discriminated. And carrying out different spatial domain noise reduction and time domain noise reduction processing on different areas of the current infrared light image by using different guiding intensities. After the time domain noise reduction and the space domain noise reduction are carried out once, the space domain noise reduction is carried out once again, so that the requirements for the time domain noise reduction and the space domain noise reduction are reduced, and the visual effect and the impression of the final image are better. Meanwhile, different fusion intensities are set according to the difference of motion information of the current visible light image and the current infrared light image in the motion area and the static area, then the fusion intensities are adjusted by utilizing the guiding intensities, more information of the current infrared light image is fused in the motion area of the current visible light noise reduction image, less information of the current infrared light image is fused in the static area, more detailed information can be supplemented in the motion area, the visual impression is better, the previous information can be fully utilized by reducing the fusion intensities in the static area, and the fidelity of brightness information is guaranteed.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a motion fusion noise reduction device according to an embodiment of the present disclosure. In this embodiment, the motion fusion noise reduction apparatus includes an acquisition module 61, a noise reduction module 62, and a fusion module 63.
The acquisition module 61 is used for acquiring a current visible light image and a current infrared light image of a target; the noise reduction module 62 is configured to perform time domain noise reduction on the current visible light image to obtain a current visible light noise reduction image; the fusion module 63 is configured to fuse the motion region of the current infrared light image and the motion region of the current visible light noise reduction image to obtain a current visible light fusion image corresponding to the motion region.
In the embodiment, after the spatial domain noise reduction is performed on the current visible light image, the motion region image of the current visible light noise reduction image is fused with the corresponding motion region image of the current infrared light image, so that the defect that the motion region of the current visible light image cannot supplement image details by effectively utilizing time domain information is overcome, and the detailed information of the motion region of the current visible light image is better, and the visual impression is better.
Referring to fig. 7, fig. 7 is a schematic structural diagram of a motion fusion noise reduction device according to an embodiment of the present disclosure. In this embodiment, the motion fusion noise reduction device 71 includes a processor 72.
The processor 72 may also be referred to as a CPU (Central Processing Unit). The processor 72 may be an integrated circuit chip having signal processing capabilities. The processor 72 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor 72 may be any conventional processor or the like.
The motion fusion noise reduction device 71 may further include a memory (not shown) for storing instructions and data required for the processor 72 to operate.
The processor 72 is configured to execute instructions to implement the methods provided by any of the embodiments of the motion fusion noise reduction method of the present application and any non-conflicting combinations thereof.
Referring to fig. 8, fig. 8 is a schematic structural diagram of a computer-readable storage medium according to an embodiment of the present disclosure. The computer readable storage medium 81 of the embodiments of the present application stores instructions/program data 82 that when executed enable the methods provided by any of the embodiments of the motion fusion noise reduction method of the present application, as well as any non-conflicting combinations. The instructions/program data 82 may form a program file stored in the storage medium 81 in the form of a software product, so as to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute all or part of the steps of the methods according to the embodiments of the present application. And the aforementioned storage medium 81 includes: various media capable of storing program codes, such as a usb disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, or terminal devices, such as a computer, a server, a mobile phone, and a tablet.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The above description is only an embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes performed by the content of the present specification and the attached drawings, or applied to other related technical fields directly or indirectly, are included in the scope of the present invention.
Claims (13)
1. A motion fusion noise reduction method, the method comprising:
acquiring a current visible light image and a current infrared light image of a target;
performing time domain noise reduction on the current visible light image to obtain a current visible light noise reduction image, wherein the visible light time domain noise reduction intensity of a motion region of the current visible light image is smaller than the visible light time domain noise reduction intensity of a static region of the current visible light image;
and fusing the motion area image of the current infrared light image with the motion area image of the current visible light noise reduction image to obtain a current visible light fused image of the corresponding motion area.
2. The motion fusion noise reduction method according to claim 1, wherein the temporally reducing the current visible light image comprises:
and fusing the static area image of the current infrared light image with the static area image of the current visible light noise reduction image to obtain a current visible light fused image corresponding to the static area.
3. The motion fusion noise reduction method according to claim 2, wherein the fusing the motion region image of the current infrared light image and the motion region image of the current visible light noise reduction image to obtain the current visible light fusion image of the corresponding motion region comprises:
acquiring a first guiding intensity and a second guiding intensity of the current infrared light image, wherein the first guiding intensity is greater than the second guiding intensity; the first guiding intensity is obtained from a first area image of the current infrared light image, the second guiding intensity is obtained from a second area image of the current infrared light image, and the signal-to-noise ratio of the first area image of the current infrared light image is higher than that of the second area image of the current infrared light image;
carrying out weighted fusion on the motion area image of the current infrared light image and the motion image of the current visible light noise reduction image; carrying out weighted fusion on the static area image of the current infrared light image and the static area image of the current visible light noise reduction image;
the weight of the moving area image of the current infrared image is a first fusion intensity, the weight of the still area image of the current infrared image is a second fusion intensity, the first fusion intensity and the second fusion intensity are calculated by respectively using the first guide intensity and the second guide intensity, and the first fusion intensity is greater than the second fusion intensity.
4. The motion fusion noise reduction method according to claim 3, wherein the fusing the motion region image of the current infrared light image with the motion region image of the current visible light noise reduction image comprises:
and carrying out noise reduction processing on the current infrared light image, wherein the noise reduction processing comprises time domain noise reduction and/or space domain noise reduction.
5. The motion fusion noise reduction method of claim 1, wherein the temporally denoising the current visible light image comprises:
and performing spatial domain noise reduction on the current visible light image.
6. The motion fusion noise reduction method according to claim 1, wherein the temporally reducing the current visible light image comprises:
and performing spatial domain noise reduction on the current visible light image.
7. The motion fusion noise reduction method according to claim 1, 5 or 6, wherein the temporally noise reducing the current visible light image comprises:
acquiring a plurality of historical visible light noise reduction images of the target, wherein the historical visible light noise reduction images are obtained by performing noise reduction processing on historical visible light images;
acquiring comprehensive visible light time domain noise reduction intensity by utilizing the plurality of historical visible light noise reduction images and the current visible light image;
and performing time domain noise reduction on the current visible light image by combining the comprehensive visible light time domain noise reduction intensity and the time domain noise reduction algorithm.
8. The motion fusion noise reduction method according to claim 7, wherein the performing noise reduction processing on the historical visible light image comprises:
sequentially carrying out time domain noise reduction processing and spatial domain noise reduction processing on the historical visible light image; or
And sequentially carrying out spatial domain noise reduction processing and time domain noise reduction processing on the historical visible light image.
9. The motion fusion noise reduction method of claim 7, wherein the obtaining a composite visible light temporal noise reduction intensity using the plurality of historical visible light noise reduction images and the current visible light image comprises:
acquiring infrared light time domain noise reduction intensity of the infrared light image and initial visible light time domain noise reduction intensity of the current visible light image, wherein the infrared light time domain noise reduction intensity of the infrared light image comprises a first infrared light time domain noise reduction intensity and a second infrared light time domain noise reduction intensity, the first infrared light time domain noise reduction intensity is the noise reduction intensity for performing time domain noise reduction processing on a first area image of the infrared light image, the second infrared light time domain noise reduction intensity is the noise reduction intensity for performing time domain noise reduction processing on a second area image of the infrared light image, and the signal-to-noise ratio of the first area image of the infrared light image is higher than the signal-to-noise ratio of the second area image of the infrared light image; the initial visible light time domain noise reduction intensity of the current visible light image comprises a first initial visible light time domain noise reduction intensity and a second initial visible light time domain noise reduction intensity, the first initial visible light time domain noise reduction intensity is a time domain noise reduction intensity of a first region image corresponding to the visible light image, and the second initial visible light time domain noise reduction intensity is a time domain noise reduction intensity of a second region image corresponding to the visible light image;
carrying out weighted fusion on the first infrared light time domain noise reduction intensity and the first initial visible light time domain noise reduction intensity; carrying out weighted fusion on the second infrared light time domain noise reduction intensity and the second initial visible light time domain noise reduction intensity to obtain the comprehensive visible light time domain noise reduction intensity; the weight of the first infrared optical time domain noise reduction intensity is a first guide intensity, and the weight of the second infrared optical time domain noise reduction intensity is a second guide intensity;
wherein the first guiding intensity is obtained from a first area image of the current infrared light image, and the second guiding intensity is obtained from a second area image of the current infrared light image.
10. The motion fusion noise reduction method according to claim 5 or 6, wherein the spatially reducing the noise of the current visible light image comprises:
acquiring infrared light airspace noise reduction intensity of the infrared light image and initial visible light airspace noise reduction intensity of the current visible light image, wherein the infrared light airspace noise reduction intensity of the infrared light image comprises first infrared light airspace noise reduction intensity and second infrared light airspace noise reduction intensity, the first infrared light airspace noise reduction intensity is the noise reduction intensity for performing airspace noise reduction processing on a first area image of the infrared light image, the second visible light airspace noise reduction intensity is the noise reduction intensity for performing airspace noise reduction processing on a second area image of the infrared light image, and the signal-to-noise ratio of the first area image of the infrared light image is higher than the signal-to-noise ratio of the second area image of the infrared light image; the initial visible light spatial noise reduction intensity of the current visible light image comprises a first initial visible light spatial noise reduction intensity and a second initial visible light spatial noise reduction intensity, wherein the first initial visible light spatial noise reduction intensity is the spatial noise reduction intensity of a first region image corresponding to the visible light image, and the second initial visible light spatial noise reduction intensity is the spatial noise reduction intensity of a second region image corresponding to the visible light image;
carrying out weighted fusion on the first infrared light spatial noise reduction intensity and the first initial visible light spatial noise reduction intensity; weighting and fusing the second infrared light airspace noise reduction intensity and the second initial visible light airspace noise reduction intensity to obtain comprehensive visible light airspace noise reduction intensity; the weight of the first infrared airspace noise reduction intensity is a first guide intensity, and the weight of the second infrared airspace noise reduction intensity is a second guide intensity;
wherein the first guiding intensity is obtained from a first area image of the current infrared light image, and the second guiding intensity is obtained from a second area image of the current infrared light image;
and performing spatial domain noise reduction on the current visible light image by combining the comprehensive visible light spatial domain noise reduction intensity and a spatial domain noise reduction algorithm.
11. The method of claim 1, wherein the fusing the current infrared light image of the motion region with the current visible light noise-reduced image of the motion region to obtain the current visible light fused image of the corresponding motion region further comprises:
and performing spatial domain noise reduction on the visible light noise reduction image.
12. A motion fusion noise reduction device comprising a processor for executing instructions to implement the motion fusion noise reduction method according to any of claims 1-11.
13. A computer-readable storage medium for storing instructions/program data executable to implement a motion fusion noise reduction method according to any one of claims 1-11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110604899.8A CN113538255A (en) | 2021-05-31 | 2021-05-31 | Motion fusion noise reduction method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110604899.8A CN113538255A (en) | 2021-05-31 | 2021-05-31 | Motion fusion noise reduction method and device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113538255A true CN113538255A (en) | 2021-10-22 |
Family
ID=78124537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110604899.8A Pending CN113538255A (en) | 2021-05-31 | 2021-05-31 | Motion fusion noise reduction method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113538255A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102769722A (en) * | 2012-07-20 | 2012-11-07 | 上海富瀚微电子有限公司 | Time-space domain hybrid video noise reduction device and method |
| CN102982518A (en) * | 2012-11-06 | 2013-03-20 | 扬州万方电子技术有限责任公司 | Fusion method of infrared image and visible light dynamic image and fusion device of infrared image and visible light dynamic image |
| 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 |
| KR20150024169A (en) * | 2013-08-26 | 2015-03-06 | 삼성테크윈 주식회사 | Image fusion system and image fusion method |
| 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 |
| CN110490811A (en) * | 2019-05-31 | 2019-11-22 | 杭州海康威视数字技术股份有限公司 | Image noise reduction apparatus and image denoising method |
| WO2020238905A1 (en) * | 2019-05-31 | 2020-12-03 | 杭州海康威视数字技术股份有限公司 | Image fusion device and method |
| CN112435183A (en) * | 2020-11-17 | 2021-03-02 | 浙江大华技术股份有限公司 | Image noise reduction method and device and storage medium |
| CN112866506A (en) * | 2019-11-08 | 2021-05-28 | 海信视像科技股份有限公司 | Time domain noise reduction method and device for video image and storage medium |
-
2021
- 2021-05-31 CN CN202110604899.8A patent/CN113538255A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102769722A (en) * | 2012-07-20 | 2012-11-07 | 上海富瀚微电子有限公司 | Time-space domain hybrid video noise reduction device and method |
| CN102982518A (en) * | 2012-11-06 | 2013-03-20 | 扬州万方电子技术有限责任公司 | Fusion method of infrared image and visible light dynamic image and fusion device of infrared image and visible light dynamic image |
| 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 |
| KR20150024169A (en) * | 2013-08-26 | 2015-03-06 | 삼성테크윈 주식회사 | Image fusion system and image fusion method |
| 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 |
| CN110490811A (en) * | 2019-05-31 | 2019-11-22 | 杭州海康威视数字技术股份有限公司 | Image noise reduction apparatus and image denoising method |
| WO2020238970A1 (en) * | 2019-05-31 | 2020-12-03 | 杭州海康威视数字技术股份有限公司 | Image denoising device and image denoising method |
| WO2020238905A1 (en) * | 2019-05-31 | 2020-12-03 | 杭州海康威视数字技术股份有限公司 | Image fusion device and method |
| CN112866506A (en) * | 2019-11-08 | 2021-05-28 | 海信视像科技股份有限公司 | Time domain noise reduction method and device for video image and storage medium |
| CN112435183A (en) * | 2020-11-17 | 2021-03-02 | 浙江大华技术股份有限公司 | Image noise reduction method and device and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| DAVID EDUARDO MORENO-VILLAMARÍN;: "David Eduardo Moreno-Villamarín;", IEEE TRANSACTIONS ON IMAGE PROCESSING, 19 April 2017 (2017-04-19), pages 1 - 13 * |
| 韩亦勇: "红外与可见光图像融合技术研究", 中国硕士学位论文全文数据库信息科技辑, 15 May 2012 (2012-05-15), pages 1 - 13 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9934438B2 (en) | Scene recognition method and apparatus | |
| EP2987135B1 (en) | Reference image selection for motion ghost filtering | |
| CN102077572B (en) | Method and apparatus for motion blur and ghosting prevention in imaging system | |
| US9202263B2 (en) | System and method for spatio video image enhancement | |
| US8149336B2 (en) | Method for digital noise reduction in low light video | |
| JP7077395B2 (en) | Multiplexed high dynamic range image | |
| US8582915B2 (en) | Image enhancement for challenging lighting conditions | |
| Kim et al. | A novel approach for denoising and enhancement of extremely low-light video | |
| US10257449B2 (en) | Pre-processing for video noise reduction | |
| Ko et al. | Artifact-free low-light video enhancement using temporal similarity and guide map | |
| CN110445951B (en) | Video filtering method and device, storage medium and electronic device | |
| WO2023137956A1 (en) | Image processing method and apparatus, electronic device, and storage medium | |
| CN110740266B (en) | Image frame selection method and device, storage medium and electronic equipment | |
| US10621703B2 (en) | Image processing apparatus, image processing method, and program | |
| Jia et al. | Real-time content adaptive contrast enhancement for see-through fog and rain | |
| US20170069068A1 (en) | Wide dynamic range imaging method | |
| Wang et al. | Low-light image enhancement based on virtual exposure | |
| CN112950502B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN110136085B (en) | Image noise reduction method and device | |
| CN116468636A (en) | Low-illumination enhancement method, device, electronic equipment and readable storage medium | |
| CN109982012B (en) | Image processing method and device, storage medium and terminal | |
| CN113538255A (en) | Motion fusion noise reduction method and device and computer readable storage medium | |
| Guthier et al. | Algorithms for a real-time HDR video system | |
| CN113259594A (en) | Image processing method and device, computer readable storage medium and terminal | |
| Nokovic et al. | Image enhancement by jetson tx2 embedded ai computing device |
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 |