CN110689496A - Method and device for determining noise reduction model, electronic equipment and computer storage medium - Google Patents

Method and device for determining noise reduction model, electronic equipment and computer storage medium Download PDF

Info

Publication number
CN110689496A
CN110689496A CN201910915975.XA CN201910915975A CN110689496A CN 110689496 A CN110689496 A CN 110689496A CN 201910915975 A CN201910915975 A CN 201910915975A CN 110689496 A CN110689496 A CN 110689496A
Authority
CN
China
Prior art keywords
noise reduction
denoised
image
image block
noise
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.)
Granted
Application number
CN201910915975.XA
Other languages
Chinese (zh)
Other versions
CN110689496B (en
Inventor
周舒畅
张继栋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Aixin Yuanzhi Semiconductor Co ltd
Original Assignee
Beijing Maigewei Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Maigewei Technology Co Ltd filed Critical Beijing Maigewei Technology Co Ltd
Priority to CN201910915975.XA priority Critical patent/CN110689496B/en
Publication of CN110689496A publication Critical patent/CN110689496A/en
Application granted granted Critical
Publication of CN110689496B publication Critical patent/CN110689496B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • 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
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Picture Signal Circuits (AREA)

Abstract

The invention provides a method and a device for determining a noise reduction model, electronic equipment and a computer storage medium, wherein the method comprises the following steps: acquiring a video stream to be denoised; in a plurality of frames of images to be denoised, determining the noise level and the image attribute of a current image block to be denoised in a current frame of image to be denoised; and determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction, and performing noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model. The method can provide a proper target noise reduction model for each image block to be subjected to noise reduction in the video stream to be subjected to noise reduction, so that the calculated amount can be reduced to the maximum extent and the power consumption can be reduced when the image frames in the video stream are subjected to noise reduction, and the technical problems of the waste of the calculated amount and the serious power consumption when the noise reduction model determined in the prior art is used for performing noise reduction on the image frames in the video stream are solved.

Description

Method and device for determining noise reduction model, electronic equipment and computer storage medium
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a method and an apparatus for determining a noise reduction model, an electronic device, and a computer storage medium.
Background
Image noise reduction is a very important part of the ISP (Image Signal Processing) flow, and the noise level of an Image is strongly correlated with the current imaging parameters (e.g., ISO (sensitivity), exposure time, whether infrared filtering is turned on, ambient lighting).
In the prior art, when denoising image frames in a video stream, each image frame in the video stream is generally denoised by a certain denoising model. In order to ensure the noise reduction effect of all image frames, the calculation amount of the noise reduction model is often large, and actually, for some image frames (for example, background images), the noise reduction model with the large calculation amount is not needed at all when noise reduction is performed, so that a good noise reduction effect can be achieved.
In summary, when the noise reduction model determined in the prior art performs noise reduction processing on image frames in a video stream, there are technical problems of wasted computation amount and serious power consumption.
Disclosure of Invention
In view of the above, the present invention provides a method and an apparatus for determining a noise reduction model, an electronic device, and a computer storage medium, so as to alleviate the technical problems of wasted computation and severe power consumption when the noise reduction model determined in the prior art performs noise reduction processing on image frames in a video stream.
In a first aspect, an embodiment of the present invention provides a method for determining a noise reduction model, including: acquiring a video stream to be denoised; the video stream to be denoised comprises a plurality of frames of images to be denoised, and each frame of image to be denoised comprises a plurality of image blocks to be denoised; in a plurality of frames of images to be denoised, determining the noise level and the image attribute of a current image block to be denoised in a current frame of image to be denoised; the image attribute includes any one of: a foreground attribute and a background attribute; and determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction, and performing noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model.
Further, when the target noise reduction model is determined based on the noise level and the image attribute of the current image block to be noise-reduced, in the case that the target noise reduction model is used to perform noise reduction processing on the image block to be noise-reduced corresponding to the position of the current image block to be noise-reduced in the next frame of image to be noise-reduced, the method further includes: performing noise reduction processing on the current image block to be subjected to noise reduction based on a target noise reduction model determined by the noise level and the image attribute of the image block to be subjected to noise reduction; and the target image block to be denoised is an image block to be denoised in the previous frame of image to be denoised, which corresponds to the position of the current image block to be denoised.
Further, under the condition that the target noise reduction model is used for performing noise reduction processing on the image block to be noise reduced corresponding to the position of the current image block to be noise reduced in the next frame of image to be noise reduced, and when the current frame of image to be noise reduced is the first frame of image to be noise reduced or the second frame of image to be noise reduced in the video stream to be noise reduced, the method further comprises the following steps: and carrying out noise reduction treatment on each image block to be subjected to noise reduction in the first frame image to be subjected to noise reduction or the second frame image to be subjected to noise reduction by adopting a preset initial noise reduction model.
Further, under the condition that the target noise reduction model is used for performing noise reduction processing on the current image block to be subjected to noise reduction, and when the current image block to be subjected to noise reduction is the first image block to be subjected to noise reduction in the video stream to be subjected to noise reduction, the method further includes: and denoising each image block to be denoised in the first frame image to be denoised by adopting a preset initial denoising model.
Further, in the multiple frames of images to be denoised, the step of determining the noise level and the image attribute of the current image block to be denoised in the current frame of image to be denoised comprises: adopting a noise level prediction model to carry out noise level prediction on the current image block to be denoised to obtain the noise level of the current image block to be denoised; and determining the image attribute of the current image block to be denoised based on the current image block to be denoised and the target image block to be denoised.
Further, the step of determining the image attribute of the current image block to be denoised based on the current image block to be denoised and the target image block to be denoised comprises: calculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised according to the pixel values of corresponding pixel points in the current image block to be denoised and the target image block to be denoised; and determining the image attribute of the current image block to be denoised according to the pixel value variation degree.
Further, the step of calculating the degree of change of the pixel value of the current image block to be denoised relative to the target image block to be denoised according to the pixel values of the corresponding pixel points in the current image block to be denoised and the target image block to be denoised comprises: calculation formula according to pixel value variation degree
Figure BDA0002215096150000031
Calculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised; c represents the degree of change of the pixel value,
Figure BDA0002215096150000032
representing the pixel value of the jth pixel point of the current image block i to be denoised,
Figure BDA0002215096150000033
and N represents the total number of pixel points in the current image block to be denoised.
Further, the step of determining the image attribute of the current image block to be denoised according to the pixel value variation degree includes: if the distance between the pixel value variation degree and a first preset threshold is smaller than the distance between the pixel value variation degree and a second preset threshold, determining the image attribute of the current image block to be denoised as a background attribute; the first preset threshold is smaller than the second preset threshold; and if the distance between the pixel value variation degree and the first preset threshold is greater than the distance between the pixel value variation degree and the second preset threshold, determining the image attribute of the current image block to be denoised as a foreground attribute.
Further, the step of determining a target noise reduction model based on the noise level and the image property of the current image block to be noise reduced includes: and searching a noise reduction model corresponding to the noise level and the image attribute of the current image block to be subjected to noise reduction in a preset noise reduction model lookup table, and taking the searched noise reduction model as the target noise reduction model.
Further, the preset denoising model lookup table is: a table of correspondence between noise levels and image attributes and noise reduction models; and training an original noise reduction model by using image samples to be noise reduced with the same noise level and the same image attribute in advance to obtain a corresponding relation table between the noise level and the image attribute and the noise reduction model.
Further, the method further comprises: acquiring a noise level image sample set; included in the set of noise level image samples are: noise levels corresponding to the multiple noise level image samples and each noise level image sample; and training an original noise level prediction model based on the noise level image sample set to obtain the noise level prediction model.
In a second aspect, an embodiment of the present invention further provides a device for determining a noise reduction model, including: the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a video stream to be denoised; the video stream to be denoised comprises a plurality of frames of images to be denoised, and each frame of image to be denoised comprises a plurality of image blocks to be denoised; the first determining unit is used for determining the noise level and the image attribute of the current image block to be denoised in the current frame image to be denoised in a plurality of frames of images to be denoised; the image attribute includes any one of: a foreground attribute and a background attribute; and the second determining unit is used for determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction so as to perform noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction through the target noise reduction model.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method according to any one of the above first aspects when executing the computer program.
In a fourth aspect, an embodiment of the present invention provides a computer storage medium, on which a computer program is stored, and when the computer program runs on a computer, the computer executes the steps of executing the method of any one of the first aspect.
In the embodiment of the invention, a video stream to be denoised is obtained firstly; then, in a plurality of frames of images to be denoised of a video stream to be denoised, determining the noise level and the image attribute of a current image block to be denoised in a current frame of image to be denoised; and finally, determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction, and performing noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model. As can be seen from the above description, the method can determine the target noise reduction model based on the noise level and the image attribute of the current image block to be noise reduced, that is, the method comprehensively considers the noise level and the image attribute of the current image block to be noise reduced, and further determines the target noise reduction model which can not only meet the requirement of the calculated amount of the current image block to be noise reduced, but also does not cause the waste of the calculated amount, so that when the noise reduction process is performed by the target noise reduction model, the calculated amount can be saved, and the power consumption can be reduced, that is, the method can provide a proper target noise reduction model for each image block to be noise reduced in the video stream to be noise reduced, so that when the noise reduction process is performed on the image frames in the video stream, the calculated amount can be reduced to the maximum, the power consumption can be reduced, and the waste of the calculated amount when the noise reduction, the technical problem of serious power consumption.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic diagram of an electronic device according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for determining a noise reduction model according to an embodiment of the present invention;
fig. 3 is a flowchart of a method for determining a noise level and an image attribute of a current image block to be denoised in a current image to be denoised according to an embodiment of the present invention;
fig. 4 is a flowchart of a method for determining image attributes of a current image block to be denoised based on the current image block to be denoised and a target image block to be denoised according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a device for determining a noise reduction model according to an embodiment of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1:
first, an electronic device 100 for implementing an embodiment of the present invention, which may be used to run a method for determining a noise reduction model according to embodiments of the present invention, is described with reference to fig. 1.
As shown in FIG. 1, electronic device 100 includes one or more processors 102, one or more memories 104, an input device 106, an output device 108, and a camera 110, which are interconnected via a bus system 112 and/or other form of connection mechanism (not shown). It should be noted that the components and structure of the electronic device 100 shown in fig. 1 are exemplary only, and not limiting, and the electronic device may have other components and structures as desired.
The processor 102 may be implemented in at least one hardware form of a Digital Signal Processor (DSP), a Field-Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), and an asic (application Specific Integrated circuit), and the processor 102 may be a Central Processing Unit (CPU) or other form of Processing unit having data Processing capability and/or instruction execution capability, and may control other components in the electronic device 100 to perform desired functions.
The memory 104 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc. On which one or more computer program instructions may be stored that may be executed by processor 102 to implement client-side functionality (implemented by the processor) and/or other desired functionality in embodiments of the invention described below. Various applications and various data, such as various data used and/or generated by the applications, may also be stored in the computer-readable storage medium.
The input device 106 may be a device used by a user to input instructions and may include one or more of a keyboard, a mouse, a microphone, a touch screen, and the like.
The output device 108 may output various information (e.g., images or sounds) to the outside (e.g., a user), and may include one or more of a display, a speaker, and the like.
The camera 110 is configured to capture a video stream to be noise-reduced, where the video stream to be noise-reduced captured by the camera is processed by the method for determining the noise-reduction model to obtain a target noise-reduction model, for example, the camera may capture an image (e.g., a photo, a video, etc.) desired by a user, and then process the image by the method for determining the noise-reduction model to obtain the target noise-reduction model, and the camera may further store the captured image in the memory 104 for use by other components.
Exemplarily, an electronic device for implementing the method for determining a noise reduction model according to an embodiment of the present invention may be implemented as a smart mobile terminal such as a smartphone, a tablet computer, or the like, and may also be implemented as a camera device having a computing capability.
Example 2:
in accordance with an embodiment of the present invention, there is provided an embodiment of a method for determining a noise reduction model, it should be noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than here.
Fig. 2 is a flowchart of a method for determining a noise reduction model according to an embodiment of the present invention, as shown in fig. 2, the method includes the following steps:
step S202, obtaining the video stream to be denoised.
The video stream to be denoised comprises a plurality of frames of images to be denoised, and each frame of image to be denoised comprises a plurality of image blocks to be denoised.
In this embodiment of the present invention, the video stream to be noise-reduced may be a video stream obtained by real-time shooting by a camera, may also be a pre-stored video stream, and may also be a video stream obtained by downloading from a target location.
In addition, the method for determining the noise reduction model in the embodiment of the present invention may be applied to a hardware processor in a camera, and when the method for determining the noise reduction model is applied to the hardware processor in the camera, the hardware processor can process a to-be-noise-reduced image block of a to-be-noise-reduced image in a video stream captured by the camera in real time; of course, the method for determining the noise reduction model may also be used in other devices with computing capabilities (e.g., a computer, a mobile phone, a tablet computer, etc.), and when the method for determining the noise reduction model is applied in other devices with computing capabilities, the device may perform post-processing on an image block to be noise reduced of an image to be noise reduced in a video stream.
It should be noted that the multiple image blocks to be denoised in each frame of the image to be denoised are obtained by dividing each frame of the image to be denoised in advance according to an image block division strategy, and the image block division strategy may be determined according to the processing capability of the hardware device. If the processing capacity of the hardware equipment is very strong, the image blocks to be denoised obtained by division can be relatively large; and if the processing capability of the hardware equipment is poor, the image blocks to be denoised obtained by division can be smaller.
Step S204, in a plurality of frames of images to be denoised, determining the noise level and the image attribute of the current image block to be denoised in the current frame of image to be denoised.
Wherein the image attributes include any of: foreground properties and background properties.
This step will be described in detail below, and will not be described herein again.
Step S206, a target noise reduction model is determined based on the noise level and the image attribute of the current image block to be subjected to noise reduction, and the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction is subjected to noise reduction through the target noise reduction model.
Therefore, the method determines a proper target noise reduction model for each image block to be subjected to noise reduction in each frame of image to be subjected to noise reduction, and then performs noise reduction on the image block to be subjected to noise reduction through the target noise reduction model. Namely, the method adopts a block noise reduction mode, namely, a target noise reduction model is respectively and pertinently selected according to the noise level and the image attribute of each image block to be subjected to noise reduction. The reason for this is that in the same frame of image to be denoised, the noise levels and image attributes of the image blocks at different positions may be different, which may cause different target denoising models corresponding to the image blocks at different positions, and in order to reduce the amount of denoising calculation to the maximum, the inventor has designed the above-mentioned blocking denoising method.
In addition, in the embodiment of the present invention, when performing image noise reduction, two modes of image noise reduction methods are provided, one is a serial mode of image noise reduction method, and the other is a parallel mode of image noise reduction method, and the two modes of image noise reduction methods are introduced below respectively:
image noise reduction method in first and serial modes
In the mode, after a target noise reduction model is determined based on the noise level and the image attribute of the current image block to be subjected to noise reduction, the determined target noise reduction model is used for carrying out noise reduction on the current image block to be subjected to noise reduction. That is, the target noise reduction model determined according to the noise level and the image attribute of each image block to be subjected to noise reduction is used for the noise reduction processing of the image block to be subjected to noise reduction, that is, the process of determining the target noise reduction model and the process of the noise reduction processing are in a serial manner.
Image noise reduction method in two-parallel mode
In the mode, after a target noise reduction model is determined based on the noise level and the image attribute of the current image block to be subjected to noise reduction, the image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction is subjected to noise reduction through the determined target noise reduction model, while the target noise reduction model is determined based on the noise level and the image attribute of the current image block to be subjected to noise reduction, the image block to be subjected to noise reduction is subjected to noise reduction based on the target noise reduction model determined based on the noise level and the image attribute of the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the previous frame of image to be subjected to noise reduction, namely, the process. In this mode, the time for the image noise reduction processing can be saved.
The inventor thinks that the image denoising method of the parallel mode is because: in a video stream, generally within 1 second, 30 frames of images are obtained by shooting, and within 1/30, the corresponding positions of the images of two adjacent frames are very similar, in terms of image scene and image quality, so that the target noise reduction model determined based on the current image block to be noise-reduced in the current frame image to be noise-reduced and the target noise reduction model determined based on the image block to be noise-reduced in the next frame image to be noise-reduced (which refers to the position corresponding to the position of the current image to be noise-reduced) in the corresponding position are not changed much, so that the target noise reduction model determined based on the current image block to be noise-reduced in the current frame image to be noise-reduced can completely perform noise reduction processing on the image block to be noise-reduced in the corresponding position in the next frame image to be noise-reduced, and the purpose of saving image noise reduction time is achieved.
In order to highlight the correctness of the image denoising method in the parallel mode, a specific example is shown below:
supposing that 10 images to be denoised are provided, the numbers of the images are respectively 1, 2, 3, … and 10, 10 image blocks to be denoised are provided in each image to be denoised, the numbers are respectively X1, X2, X3, … and X10, the image blocks to be denoised in the 1 st image to be denoised adopt a preset initial denoising model (because the image attributes of the image blocks to be denoised in the 1 st image to be denoised cannot be determined, the preset initial denoising model is adopted by default for the image blocks to be denoised in the 1 st image to be denoised), the target denoising model determined according to the noise level and the image attributes of the image blocks to be denoised in the 2 nd to 5 th images to be denoised is an a denoising model, and the target denoising model determined according to the noise level and the image attributes of the image blocks to be denoised in the next 5 images to be denoised is a model.
Then, when the image in the serial mode is denoised, an X1 image block to be denoised in the 1 st image to be denoised adopts a preset initial denoising model, an X1 image block to be denoised in the 2 nd to 5 th images to be denoised adopts an a denoising model, and an X1 image block to be denoised in the last 5 images to be denoised adopts a b denoising model;
when an image in the parallel mode is denoised, an X1 image block to be denoised in the 1 st image to be denoised adopts a preset initial denoising model, an X1 image block to be denoised in the 2 nd image to be denoised also adopts a preset initial denoising model (because a target denoising model cannot be determined according to an X1 image block to be denoised in the 1 st image to be denoised and an X1 image block in the default 2 image to be denoised also adopts a preset initial denoising model), an X1 image block in the 3 rd to 6 th image to be denoised adopts a target denoising model, namely an a denoising model, determined according to the noise level and the image attribute of an X1 image block to be denoised in the 2 th to 5 th image to be denoised, and an X1 image block in the 7 th to 10 image to be denoised adopts a b denoising model;
compared with the image noise reduction in the serial mode, the noise reduction in the parallel mode is realized by the noise reduction results in the two modes, only the noise reduction models adopted by the image block to be subjected to noise reduction in the X1 image block in the 2 nd image to be subjected to noise reduction and the image block to be subjected to noise reduction in the X1 image block in the 6 th image to be subjected to noise reduction are not the noise reduction models corresponding to the noise reduction models, but for a video stream containing a large number of image frames, the noise reduction errors are not influenced at all, and in addition, the noise reduction method in the parallel mode obviously sacrifices a small part of noise reduction precision in exchange for the effect of saving a large amount of noise reduction time.
In the embodiment of the invention, a video stream to be denoised is obtained firstly; then, in a plurality of frames of images to be denoised of a video stream to be denoised, determining the noise level and the image attribute of a current image block to be denoised in a current frame of image to be denoised; and finally, determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction, and performing noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model. As can be seen from the above description, the method can determine the target noise reduction model based on the noise level and the image attribute of the current image block to be noise reduced, that is, the method comprehensively considers the noise level and the image attribute of the current image block to be noise reduced, and further determines the target noise reduction model which can not only meet the requirement of the calculated amount of the current image block to be noise reduced, but also does not cause the waste of the calculated amount, so that when the noise reduction process is performed by the target noise reduction model, the calculated amount can be saved, and the power consumption can be reduced, that is, the method can provide a proper target noise reduction model for each image block to be noise reduced in the video stream to be noise reduced, so that when the noise reduction process is performed on the image frames in the video stream, the calculated amount can be reduced to the maximum, the power consumption can be reduced, and the waste of the calculated amount when the noise reduction, the technical problem of serious power consumption.
The above description briefly describes the method for determining the noise reduction model of the present invention, and the following description details other matters involved therein.
In an optional embodiment of the present invention, in a case that a noise reduction process is performed on an image block to be noise-reduced corresponding to a position of a current image block to be noise-reduced in a next frame of image to be noise-reduced by using a target noise reduction model, when the target noise reduction model is determined based on a noise level and an image attribute of the current image block to be noise-reduced, the method further includes:
and performing noise reduction processing on the current image block to be subjected to noise reduction based on the noise level of the image block to be subjected to noise reduction and the target noise reduction model determined by the image attribute.
And the target image block to be denoised is an image block to be denoised corresponding to the position of the current image block to be denoised in the previous frame of image to be denoised.
The process may refer to the description of 10 images to be denoised below step S206, and will not be described herein again.
In an optional embodiment of the present invention, in a case that a noise reduction processing is performed on an image block to be noise-reduced corresponding to a position of a current image block to be noise-reduced in a next image to be noise-reduced through a target noise reduction model, and when the current image to be noise-reduced is a first image to be noise-reduced or a second image to be noise-reduced in a video stream to be noise-reduced, the method further includes:
and carrying out noise reduction treatment on each image block to be subjected to noise reduction in the first frame image to be subjected to noise reduction or the second frame image to be subjected to noise reduction by adopting a preset initial noise reduction model.
Similarly, the process may refer to the description of 10 images to be denoised below step S206, which is not described herein again.
In an optional embodiment of the present invention, in a case that the target noise reduction model is used to perform noise reduction processing on a current image block to be noise reduced, and when the current image block to be noise reduced is a first image block to be noise reduced in a video stream to be noise reduced, the method further includes:
and carrying out noise reduction treatment on each image block to be subjected to noise reduction in the first frame image to be subjected to noise reduction by adopting a preset initial noise reduction model.
Similarly, the process may refer to the description of 10 images to be denoised below step S206, which is not described herein again.
The above describes the overall process of image denoising, and the following describes the details of the process:
in an alternative embodiment of the present invention, referring to fig. 3, in a plurality of frames of images to be denoised, the step of determining the noise level and the image attribute of the current image block to be denoised in the current image to be denoised comprises:
step S301, a noise level prediction model is adopted to carry out noise level prediction on the current image block to be denoised, and the noise level of the current image block to be denoised is obtained.
The noise level prediction model is a model obtained by training in advance. The specific training process is as follows:
(1) acquiring a noise level image sample set; wherein the noise level image sample set comprises: noise levels corresponding to the multiple noise level image samples and each noise level image sample;
(2) and training the original noise level prediction model based on the noise level image sample set to obtain the noise level prediction model.
When the original noise level prediction model is trained, noise level image samples are input into the original noise level prediction model, the noise level corresponding to the noise level image samples is output, the noise level output by the model is compared with the noise level corresponding to the noise level image samples in the noise level image sample set, if the noise level output by the model is not consistent with the noise level corresponding to the noise level image samples in the same noise level image sample, the original noise level prediction model is adjusted, and therefore after multiple times of adjustment, the noise level prediction model is trained.
Step S302, determining the image attribute of the current image block to be denoised based on the current image block to be denoised and the target image block to be denoised.
Referring to fig. 4, the step of determining the image attribute of the current image block to be denoised based on the current image block to be denoised and the target image block to be denoised includes:
step S401, calculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised according to the pixel values of the corresponding pixel points in the current image block to be denoised and the target image block to be denoised.
Specifically, the calculation formula is calculated according to the pixel value variation degreeCalculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised; c represents a degree of change in the pixel value,
Figure BDA0002215096150000143
representing the pixel value of the jth pixel point of the current image block i to be denoised,
Figure BDA0002215096150000142
and (3) representing the pixel value of the jth pixel point of the target image block i-1 to be denoised, wherein N represents the total number of pixel points in the current image block to be denoised.
And the target image block to be denoised is an image block to be denoised corresponding to the position of the current image block to be denoised in the previous frame of image to be denoised.
Step S402, determining the image attribute of the current image block to be denoised according to the pixel value variation degree.
Specifically, the method comprises the following steps: if the distance between the pixel value variation degree and the first preset threshold is smaller than the distance between the pixel value variation degree and the second preset threshold, determining the image attribute of the current image block to be denoised as the background attribute; and if the distance between the pixel value variation degree and the first preset threshold is greater than the distance between the pixel value variation degree and the second preset threshold, determining the image attribute of the current image block to be denoised as the foreground attribute.
The first preset threshold is smaller than the second preset threshold. The first preset threshold may be 0, and the second preset threshold may be 1, and the first preset threshold and the second preset threshold are not limited in this embodiment.
The following describes in detail the process of determining a target noise reduction model:
in an optional embodiment of the present invention, the step of determining the target noise reduction model based on the noise level and the image property of the current image block to be noise reduced includes:
and searching a noise reduction model corresponding to the noise level and the image attribute of the current image block to be subjected to noise reduction in a preset noise reduction model lookup table, and taking the searched noise reduction model as a target noise reduction model.
Wherein, the preset denoising model lookup table is as follows: and a table of correspondence between noise levels and image attributes and noise reduction models.
Specifically, after an original noise reduction model is trained in advance through image samples to be noise reduced with the same noise level and the same image attribute, a correspondence table between the noise level and the image attribute and the noise reduction model is obtained.
For example, the image samples to be denoised are foreground images with the noise level of 2, and after the original denoising model is trained by the image samples to be denoised, the obtained denoising model is the corresponding denoising model with the noise level of 2 and the image attribute of the foreground attribute.
In order to better understand the above image denoising method of the present invention, the following describes an image denoising method in a parallel mode in the form of pseudo code:
input set frames { frame0, frame1, frame2, … } (frame0 { patch0, patch1,. }.), dense network set N { N0, N1, N2, … }; v/video stream frames, where each frame is made up of a series of latches (block de-noising), a set of de-noising networks N;
denoise _ network ═ n0, n0, …, n 0; v/initialize the noise reduction network to n 0;
for i in len (frames)// traverse each frame image;
for j in len (frames [ i ])// traverse each patch of the frame;
processing the image by Res (Denoise _ network [ j ] (frames [ i ] [ j ])// noise _ network [ j ] predicted by the patch corresponding to the previous frame to obtain a denoised image Res;
noise level of a block at the same position in a next frame is predicted by a block of a current frame;
is _ back _ group Is 0else 1if threshold (| frame [ i ] [ j ] -frame [ i-1] [ j ] |)// judging whether the current frame Is the background or not according to the difference between the current frame and the previous frame;
and obtaining a Noise reduction network at the corresponding position of the next frame through table look-up.
The method for determining the noise reduction model can provide a target noise reduction model which can meet the requirement of the calculated amount and can not cause the waste of the calculated amount for each image block to be subjected to noise reduction in each frame of image to be subjected to noise reduction in a video stream, so that the calculated amount can be reduced to the maximum extent and the power consumption can be reduced when the noise reduction is carried out, and in addition, the image noise reduction method in the parallel mode can also improve the parallelism and save the time for reducing the noise of the image greatly.
Example 3:
the embodiment of the present invention further provides a device for determining a noise reduction model, where the device for determining a noise reduction model is mainly used to execute the method for determining a noise reduction model provided in the foregoing content of the embodiment of the present invention, and the device for determining a noise reduction model provided in the embodiment of the present invention is described in detail below.
Fig. 5 is a schematic diagram of a device for determining a noise reduction model according to an embodiment of the present invention, as shown in fig. 5, the device for determining a noise reduction model mainly includes: an acquisition unit 10, a first determination unit 20 and a second determination unit 30, wherein:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a video stream to be denoised; the video stream to be denoised comprises a plurality of frames of images to be denoised, and each frame of image to be denoised comprises a plurality of image blocks to be denoised;
the first determining unit is used for determining the noise level and the image attribute of the current image block to be denoised in the current frame image to be denoised in a plurality of frames of images to be denoised; the image attributes include any of: a foreground attribute and a background attribute;
and the second determining unit is used for determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction so as to perform noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model.
In the embodiment of the invention, a video stream to be denoised is obtained firstly; then, in a plurality of frames of images to be denoised of a video stream to be denoised, determining the noise level and the image attribute of a current image block to be denoised in a current frame of image to be denoised; and finally, determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction, and performing noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model. As can be seen from the above description, the method can determine the target noise reduction model based on the noise level and the image attribute of the current image block to be noise reduced, that is, the method comprehensively considers the noise level and the image attribute of the current image block to be noise reduced, and further determines the target noise reduction model which can not only meet the requirement of the calculated amount of the current image block to be noise reduced, but also does not cause the waste of the calculated amount, so that when the noise reduction process is performed by the target noise reduction model, the calculated amount can be saved, and the power consumption can be reduced, that is, the method can provide a proper target noise reduction model for each image block to be noise reduced in the video stream to be noise reduced, so that when the noise reduction process is performed on the image frames in the video stream, the calculated amount can be reduced to the maximum, the power consumption can be reduced, and the waste of the calculated amount when the noise reduction, the technical problem of serious power consumption.
Optionally, in a case that the target noise reduction model is used to perform noise reduction processing on an image block to be noise-reduced corresponding to the position of the current image block to be noise-reduced in the next frame of image to be noise-reduced, when the target noise reduction model is determined based on the noise level and the image attribute of the current image block to be noise-reduced, the apparatus is further configured to: performing noise reduction processing on the current image block to be subjected to noise reduction based on a target noise reduction model determined by the noise level and the image attribute of the image block to be subjected to noise reduction; and the target image block to be denoised is an image block to be denoised in the previous frame of image to be denoised, which corresponds to the position of the current image block to be denoised.
Optionally, in a case that the target noise reduction model is used to perform noise reduction processing on an image block to be noise-reduced corresponding to the position of the current image block to be noise-reduced in the next frame of image to be noise-reduced, and when the current frame of image to be noise-reduced is the first frame of image to be noise-reduced or the second frame of image to be noise-reduced in the video stream to be noise-reduced, the apparatus is further configured to: and carrying out noise reduction treatment on each image block to be subjected to noise reduction in the first frame image to be subjected to noise reduction or the second frame image to be subjected to noise reduction by adopting a preset initial noise reduction model.
Optionally, in a case that the target noise reduction model is used to perform noise reduction processing on the current image block to be noise reduced, and when the current image to be noise reduced is the first image to be noise reduced in the video stream to be noise reduced, the apparatus is further configured to: and carrying out noise reduction treatment on each image block to be subjected to noise reduction in the first frame image to be subjected to noise reduction by adopting a preset initial noise reduction model.
Optionally, the first determining unit is further configured to: adopting a noise level prediction model to carry out noise level prediction on the current image block to be denoised to obtain the noise level of the current image block to be denoised; and determining the image attribute of the current image block to be denoised based on the current image block to be denoised and the target image block to be denoised.
Optionally, the first determining unit is further configured to: calculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised according to the pixel values of corresponding pixel points in the current image block to be denoised and the target image block to be denoised; and determining the image attribute of the current image block to be denoised according to the pixel value variation degree.
Optionally, the first determining unit is further configured to: calculation formula according to pixel value variation degreeCalculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised; c represents a degree of change in the pixel value,
Figure BDA0002215096150000182
representing the pixel value of the jth pixel point of the current image block i to be denoised,
Figure BDA0002215096150000183
and (3) representing the pixel value of the jth pixel point of the target image block i-1 to be denoised, wherein N represents the total number of pixel points in the current image block to be denoised.
Optionally, the first determining unit is further configured to: if the distance between the pixel value variation degree and the first preset threshold is smaller than the distance between the pixel value variation degree and the second preset threshold, determining the image attribute of the current image block to be denoised as the background attribute; the first preset threshold value is smaller than the second preset threshold value; and if the distance between the pixel value variation degree and the first preset threshold is greater than the distance between the pixel value variation degree and the second preset threshold, determining the image attribute of the current image block to be denoised as the foreground attribute.
Optionally, the second determining unit is further configured to: and searching a noise reduction model corresponding to the noise level and the image attribute of the current image block to be subjected to noise reduction in a preset noise reduction model lookup table, and taking the searched noise reduction model as a target noise reduction model.
Optionally, the preset denoising model lookup table is: a table of correspondence between noise levels and image attributes and noise reduction models; and training the original noise reduction model by using the image samples to be noise reduced with the same noise level and the same image attribute in advance to obtain a corresponding relation table between the noise level and the image attribute and the noise reduction model.
Optionally, the apparatus is further configured to: acquiring a noise level image sample set; the noise level image sample set includes: noise levels corresponding to the multiple noise level image samples and each noise level image sample; and training the original noise level prediction model based on the noise level image sample set to obtain the noise level prediction model.
The device provided by the embodiment of the present invention has the same implementation principle and technical effect as the method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the method embodiments without reference to the device embodiments.
In another implementation of the present invention, there is further provided a computer storage medium having a computer program stored thereon, the computer program, when executed by a computer, performing the steps of the method of any one of the above method embodiments 2.
In addition, in the description of the embodiments of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
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 ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and 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 of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention 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 functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (14)

1. A method for determining a noise reduction model, comprising:
acquiring a video stream to be denoised; the video stream to be denoised comprises a plurality of frames of images to be denoised, and each frame of image to be denoised comprises a plurality of image blocks to be denoised;
in a plurality of frames of images to be denoised, determining the noise level and the image attribute of a current image block to be denoised in a current frame of image to be denoised; the image attribute includes any one of: a foreground attribute and a background attribute;
and determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction, and performing noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model.
2. The method according to claim 1, wherein in a case where an image block to be noise-reduced in a next frame image to be noise-reduced corresponding to the position of the current image block to be noise-reduced is subjected to noise reduction processing by the target noise reduction model, when the target noise reduction model is determined based on the noise level and the image property of the current image block to be noise-reduced, the method further comprises:
performing noise reduction processing on the current image block to be subjected to noise reduction based on a target noise reduction model determined by the noise level and the image attribute of the image block to be subjected to noise reduction; and the target image block to be denoised is an image block to be denoised in the previous frame of image to be denoised, which corresponds to the position of the current image block to be denoised.
3. The method according to claim 1, wherein in a case that the target noise reduction model is used to perform noise reduction on an image block to be noise-reduced in a next frame of image to be noise-reduced corresponding to the position of the current image block to be noise-reduced, and when the current frame of image to be noise-reduced is a first frame of image to be noise-reduced or a second frame of image to be noise-reduced in the video stream to be noise-reduced, the method further comprises:
and carrying out noise reduction treatment on each image block to be subjected to noise reduction in the first frame image to be subjected to noise reduction or the second frame image to be subjected to noise reduction by adopting a preset initial noise reduction model.
4. The method according to claim 1, wherein in a case that the current image block to be noise-reduced is subjected to noise reduction processing by the target noise reduction model, and when the current image block to be noise-reduced is a first image block to be noise-reduced in the video stream to be noise-reduced, the method further comprises:
and denoising each image block to be denoised in the first frame image to be denoised by adopting a preset initial denoising model.
5. The method according to claim 2, wherein the step of determining the noise level and the image property of the current image block to be denoised in the current image to be denoised comprises:
adopting a noise level prediction model to carry out noise level prediction on the current image block to be denoised to obtain the noise level of the current image block to be denoised;
and determining the image attribute of the current image block to be denoised based on the current image block to be denoised and the target image block to be denoised.
6. The method according to claim 5, wherein the step of determining the image attributes of the current image block to be denoised based on the current image block to be denoised and the target image block to be denoised comprises:
calculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised according to the pixel values of corresponding pixel points in the current image block to be denoised and the target image block to be denoised;
and determining the image attribute of the current image block to be denoised according to the pixel value variation degree.
7. The method according to claim 6, wherein the step of calculating the degree of change of the pixel value of the current image block to be denoised relative to the target image block to be denoised according to the pixel values of the corresponding pixel points in the current image block to be denoised and the target image block to be denoised comprises:
calculation formula according to pixel value variation degree
Figure FDA0002215096140000021
Calculating the pixel value variation degree of the current image block to be denoised relative to the target image block to be denoised; c represents the degree of change of the pixel value,
Figure FDA0002215096140000022
representing the pixel value of the jth pixel point of the current image block i to be denoised,
Figure FDA0002215096140000023
and N represents the total number of pixel points in the current image block to be denoised.
8. The method according to claim 6, wherein the step of determining the image attributes of the current image block to be denoised according to the pixel value variation degree comprises:
if the distance between the pixel value variation degree and a first preset threshold is smaller than the distance between the pixel value variation degree and a second preset threshold, determining the image attribute of the current image block to be denoised as a background attribute; the first preset threshold is smaller than the second preset threshold;
and if the distance between the pixel value variation degree and the first preset threshold is greater than the distance between the pixel value variation degree and the second preset threshold, determining the image attribute of the current image block to be denoised as a foreground attribute.
9. The method of claim 1, wherein the step of determining a target noise reduction model based on the noise level and the image properties of the current image block to be noise reduced comprises:
and searching a noise reduction model corresponding to the noise level and the image attribute of the current image block to be subjected to noise reduction in a preset noise reduction model lookup table, and taking the searched noise reduction model as the target noise reduction model.
10. The method of claim 9, wherein the predetermined noise reduction model lookup table is: a table of correspondence between noise levels and image attributes and noise reduction models; and training an original noise reduction model by using image samples to be noise reduced with the same noise level and the same image attribute in advance to obtain a corresponding relation table between the noise level and the image attribute and the noise reduction model.
11. The method of claim 5, further comprising:
acquiring a noise level image sample set; included in the set of noise level image samples are: noise levels corresponding to the multiple noise level image samples and each noise level image sample;
and training an original noise level prediction model based on the noise level image sample set to obtain the noise level prediction model.
12. An apparatus for determining a noise reduction model, comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a video stream to be denoised; the video stream to be denoised comprises a plurality of frames of images to be denoised, and each frame of image to be denoised comprises a plurality of image blocks to be denoised;
the first determining unit is used for determining the noise level and the image attribute of the current image block to be denoised in the current frame image to be denoised in a plurality of frames of images to be denoised; the image attribute includes any one of: a foreground attribute and a background attribute;
and the second determining unit is used for determining a target noise reduction model based on the noise level and the image attribute of the current image block to be subjected to noise reduction so as to perform noise reduction processing on the image block to be subjected to noise reduction corresponding to the position of the current image block to be subjected to noise reduction in the next frame of image to be subjected to noise reduction or the current image block to be subjected to noise reduction through the target noise reduction model.
13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of the preceding claims 1 to 11 when executing the computer program.
14. A computer storage medium, having a computer program stored thereon, which, when executed by a computer, performs the steps of the method of any of claims 1 to 11.
CN201910915975.XA 2019-09-25 2019-09-25 Method and device for determining noise reduction model, electronic equipment and computer storage medium Active CN110689496B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910915975.XA CN110689496B (en) 2019-09-25 2019-09-25 Method and device for determining noise reduction model, electronic equipment and computer storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910915975.XA CN110689496B (en) 2019-09-25 2019-09-25 Method and device for determining noise reduction model, electronic equipment and computer storage medium

Publications (2)

Publication Number Publication Date
CN110689496A true CN110689496A (en) 2020-01-14
CN110689496B CN110689496B (en) 2022-10-14

Family

ID=69110328

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910915975.XA Active CN110689496B (en) 2019-09-25 2019-09-25 Method and device for determining noise reduction model, electronic equipment and computer storage medium

Country Status (1)

Country Link
CN (1) CN110689496B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111507923A (en) * 2020-04-21 2020-08-07 浙江大华技术股份有限公司 Method, device, equipment and medium for processing noise of video image
CN113536214A (en) * 2020-04-14 2021-10-22 浙江大华技术股份有限公司 Image noise reduction method and device and storage device
WO2022089079A1 (en) * 2020-10-30 2022-05-05 湖北锐世数字医学影像科技有限公司 Image processing method, apparatus and system, and device and computer-readable storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07131680A (en) * 1993-11-09 1995-05-19 Hitachi Ltd Noise reeduction circuit
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
CN103632352A (en) * 2013-11-01 2014-03-12 华为技术有限公司 Method for time domain noise reduction of noise image and related device
CN109639982A (en) * 2019-01-04 2019-04-16 Oppo广东移动通信有限公司 A kind of image denoising method, device, storage medium and terminal
CN109671026A (en) * 2018-11-28 2019-04-23 浙江大学 Gray level image noise-reduction method based on empty convolution and automatic encoding and decoding neural network
CN110248106A (en) * 2019-06-13 2019-09-17 Oppo广东移动通信有限公司 Image denoising method, device, electronic equipment and storage medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07131680A (en) * 1993-11-09 1995-05-19 Hitachi Ltd Noise reeduction circuit
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
CN103632352A (en) * 2013-11-01 2014-03-12 华为技术有限公司 Method for time domain noise reduction of noise image and related device
CN109671026A (en) * 2018-11-28 2019-04-23 浙江大学 Gray level image noise-reduction method based on empty convolution and automatic encoding and decoding neural network
CN109639982A (en) * 2019-01-04 2019-04-16 Oppo广东移动通信有限公司 A kind of image denoising method, device, storage medium and terminal
CN110248106A (en) * 2019-06-13 2019-09-17 Oppo广东移动通信有限公司 Image denoising method, device, electronic equipment and storage medium

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113536214A (en) * 2020-04-14 2021-10-22 浙江大华技术股份有限公司 Image noise reduction method and device and storage device
CN111507923A (en) * 2020-04-21 2020-08-07 浙江大华技术股份有限公司 Method, device, equipment and medium for processing noise of video image
CN111507923B (en) * 2020-04-21 2023-09-12 浙江大华技术股份有限公司 Noise processing method, device, equipment and medium for video image
WO2022089079A1 (en) * 2020-10-30 2022-05-05 湖北锐世数字医学影像科技有限公司 Image processing method, apparatus and system, and device and computer-readable storage medium

Also Published As

Publication number Publication date
CN110689496B (en) 2022-10-14

Similar Documents

Publication Publication Date Title
CN108898567B (en) Image noise reduction method, device and system
CN108961303B (en) Image processing method and device, electronic equipment and computer readable medium
CN110121882B (en) Image processing method and device
JP6742732B2 (en) Method for generating HDR image of scene based on trade-off between luminance distribution and motion
CN110689496B (en) Method and device for determining noise reduction model, electronic equipment and computer storage medium
CN108924420B (en) Image shooting method, image shooting device, image shooting medium, electronic equipment and model training method
WO2017101489A1 (en) Method and device for image filtering
WO2021114868A1 (en) Denoising method, terminal, and storage medium
CN109064504B (en) Image processing method, apparatus and computer storage medium
JP2015225665A (en) Image noise removal method and image noise removal device
US11836898B2 (en) Method and apparatus for generating image, and electronic device
CN108846814B (en) Image processing method, image processing device, readable storage medium and computer equipment
WO2017113917A1 (en) Imaging method, imaging apparatus, and terminal
WO2017185772A1 (en) Method and device for video image enhancement and computer storage medium
WO2014070489A1 (en) Recursive conditional means image denoising
CN110213479B (en) Anti-shake method and device for video shooting
JP7212554B2 (en) Information processing method, information processing device, and program
CN110475117B (en) Image compression method and device, electronic equipment and computer storage medium
CN110517201B (en) Method and device for smoothing filtering along environment-friendly edge and electronic equipment
US20150187051A1 (en) Method and apparatus for estimating image noise
CN109934789A (en) Image de-noising method, device and electronic equipment
CN111028176A (en) Fingerprint image processing method and device, electronic equipment and computer readable medium
CN112288664A (en) High dynamic range image fusion method and device and electronic equipment
CN112801890B (en) Video processing method, device and equipment
CN108734712B (en) Background segmentation method and device and computer storage medium

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
TR01 Transfer of patent right

Effective date of registration: 20230818

Address after: Room 59, 17th Floor, Science and Technology Innovation Building, No. 777 Zhongguan West Road, Zhuangshi Street, Zhenhai District, Ningbo City, Zhejiang Province, 315200

Patentee after: Aixin Yuanzhi Semiconductor (Ningbo) Co.,Ltd.

Address before: 316-318, block a, Rongke Information Center, No.2, south academy of Sciences Road, Haidian District, Beijing

Patentee before: MEGVII (BEIJING) TECHNOLOGY Co.,Ltd.

TR01 Transfer of patent right
CP03 Change of name, title or address

Address after: Room 59, 17th Floor, Science and Technology Innovation Building, No. 777 Zhongguan West Road, Zhuangshi Street, Ningbo City, Zhejiang Province, 315200

Patentee after: Aixin Yuanzhi Semiconductor Co.,Ltd.

Country or region after: China

Address before: Room 59, 17th Floor, Science and Technology Innovation Building, No. 777 Zhongguan West Road, Zhuangshi Street, Zhenhai District, Ningbo City, Zhejiang Province, 315200

Patentee before: Aixin Yuanzhi Semiconductor (Ningbo) Co.,Ltd.

Country or region before: China

CP03 Change of name, title or address