Video frame noise processing method and device

Abstract

The embodiment of the invention provides a video frame noise processing method and device. The method comprises the following steps: respectively distinguishing each pixel of a current video frame into a dynamic pixel, a weak dynamic pixel or a static pixel; determining the weak dynamic pixel as a dynamic pixel or a static pixel according to the pixel information of the neighborhood of the weak dynamic pixel and the significance information; and carrying out noise processing on the dynamic pixels and the static pixels. Through the video frame noise processing method and device, the problem that the noise processing effect is poor due to the fact that the accuracy of the judgment mode of the static pixels and the dynamic pixels is low in the prior art is solved, and then the effect that the video frame noise is effectively processed based on the accurate distinguishing result of the dynamic pixels and the static pixels is achieved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of video processing, and more specifically, to a method and device for processing video frame noise. Background Art

[0002] Video denoising technology began to develop with the rise of video recording technology, but early technologies were limited by hardware computing power and storage bottlenecks, making it difficult to achieve both high efficiency and high speed. After decades of development, most existing video denoising technologies use recursive ideas, and can be roughly divided into two categories based on whether they contain a motion compensation module: motion adaptive video denoising and motion compensated video denoising. Motion adaptive video denoising aims to improve the denoising effect in static areas and minimize blur in moving areas. The relevant technology center generally has low accuracy in dynamic pixel and static pixel judgment. Summary of the invention

[0003] The embodiments of the present application provide a method and device for processing video frame noise, so as to at least solve the problem in the related art that the noise processing effect is poor due to the low accuracy of the judgment method for static pixels and dynamic pixels.

[0004] According to one embodiment of the present application, a method for processing video frame noise is provided, comprising: distinguishing each pixel of the current video frame as a dynamic pixel, a weak dynamic pixel or a static pixel; determining the weak dynamic pixel as a dynamic pixel or a static pixel based on pixel information and significance information of the neighborhood of the weak dynamic pixel; and performing noise processing on the dynamic pixel and the static pixel.

[0005] In an exemplary embodiment, each of the pixels of the current video frame is distinguished as the dynamic pixel, the weak dynamic pixel or the static pixel, including: subtracting the pixel value of each target pixel in the current video frame from the pixel value of the target pixel in the previous frame, and comparing the absolute value of the obtained difference with a first threshold and a second threshold, wherein when the absolute value of the difference is less than the first threshold, the target pixel is the static pixel; when the absolute value of the difference is greater than or equal to the first threshold and less than the second threshold, the target pixel is the weak dynamic pixel; when the absolute value of the difference is greater than the second threshold, the target pixel is the dynamic pixel.

[0006] According to another embodiment of the present application, a video frame noise processing device is provided, including: a first setting module, used to distinguish each pixel of the current video frame as a dynamic pixel, a weak dynamic pixel or a static pixel; a second setting module, used to determine the weak dynamic pixel as a dynamic pixel or a static pixel based on the pixel information and significance information of the neighborhood of the weak dynamic pixel; a noise processing module, used to perform noise processing on the dynamic pixel and the static pixel.

[0007] In an exemplary embodiment, the first setting module also includes: a distinguishing unit, which is used to subtract the pixel value of each target pixel in the current video frame from the pixel value of the target pixel in the previous frame, and compare the absolute value of the obtained difference with the first threshold and the second threshold, wherein when the absolute value of the difference is less than the first threshold, the target pixel is the static pixel; when the absolute value of the difference is greater than or equal to the first threshold and less than the second threshold, the target pixel is the weak dynamic pixel; when the absolute value of the difference is greater than the second threshold, the target pixel is the dynamic pixel.

[0008] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when run.

[0009] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0010] Through the present application, since a secondary distinction is required when determining whether each pixel of the current video frame is a dynamic pixel or a static pixel, each pixel is first distinguished as a dynamic pixel, a weak dynamic pixel or a static pixel, and then the weak dynamic pixel is distinguished again, and the weak dynamic pixel is determined to be a static pixel or a dynamic pixel based on the pixel information and saliency information of the neighborhood of the weak dynamic pixel, and a more accurate dynamic pixel and static pixel distinction result is obtained, and then the dynamic pixel and the static pixel can be processed for noise. Therefore, the problem of poor noise processing effect due to the low accuracy of the judgment method for static pixels and dynamic pixels in the related art can be solved, and the effect of effectively processing the video frame noise after accurately judging the dynamic pixels and static pixels is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a hardware structure block diagram of a computer terminal according to a video frame noise processing method of an embodiment of the present application;

[0012] Figure 2 is a flowchart of a method for processing video frame noise according to an embodiment of the present application;

[0013] Figure 3 is a structural block diagram of a video frame noise processing device according to an embodiment of the present application;

[0014] Figure 4 is a structural block diagram of a video frame noise processing device according to yet another embodiment of the present application;

[0015] Figure 5 is a structural schematic diagram of a video frame noise processing system according to an embodiment of the present application;

[0016] Figure 6 is a schematic diagram of the effect of a saliency detection algorithm according to an embodiment of the present application;

[0017] Figure 7 is a schematic diagram of the overall decision of the motion and static decision module according to an embodiment of the present application;

[0018] Figure 8 is a schematic diagram of initialization of dynamic and static judgment according to an embodiment of the present application;

[0019] Fig. 9 It is a block diagram of a time domain fusion module according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0022] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 1 is a hardware structure block diagram of a computer terminal of a video frame noise processing method according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0023] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the video frame noise processing method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0024] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0025] Figure 2 is a flowchart of a method for processing video frame noise according to an embodiment of the present application. Figure 2 As shown, the process includes the following steps:

[0026] Step S202, classifying each pixel of the current video frame as a dynamic pixel, a weakly dynamic pixel or a static pixel;

[0027] Each pixel of the current video frame can be distinguished as a dynamic pixel, a weak dynamic pixel or a static pixel in a variety of ways. In an exemplary embodiment, the distinction is made by subtracting the pixel value of each target pixel in the previous frame from the pixel value of the target pixel in the current video frame, and comparing the absolute value of the obtained difference with a first threshold and a second threshold, wherein when the absolute value of the difference is less than the first threshold, the target pixel is the static pixel; when the absolute value of the difference is greater than or equal to the first threshold and less than the second threshold, the target pixel is the weak dynamic pixel; and when the absolute value of the difference is greater than the second threshold, the target pixel is the dynamic pixel.

[0028] Among them, weak dynamic pixels refer to pixels whose state change speed is less than a preset threshold and cannot respond to fast-moving images in real time.

[0029] Before comparing the absolute value of the pixel value of each target pixel in the current video frame minus the pixel value of the pixel in the previous frame with the first threshold and the second threshold, it also includes: setting the first threshold and the second threshold according to the noise estimation value of the current video frame and the noise estimation value of the previous frame.

[0030] Step S204, determining the weak dynamic pixel as a dynamic pixel or a static pixel according to the pixel information and saliency information of the neighborhood of the weak dynamic pixel;

[0031] Determine whether the dynamic pixel exists in the neighborhood of the weak dynamic pixel and whether the weak dynamic pixel is a significant pixel; when the dynamic pixel exists in the neighborhood of the weak dynamic pixel or the weak dynamic pixel is the significant pixel, determine that the weak dynamic pixel is the dynamic pixel; when there is no dynamic pixel in the neighborhood of the weak dynamic pixel and the weak dynamic pixel is not a significant pixel, determine that the weak dynamic pixel is the static pixel; wherein the neighborhood of the weak dynamic pixel is adjusted by a preset parameter.

[0032] There are many ways to determine whether a weak dynamic pixel is a significant pixel. In an exemplary embodiment of the present application, a saliency algorithm based on histogram contrast (HC) is used for saliency detection. First, the grayscale histogram of the image is calculated, and then the saliency of each pixel in the image is evaluated based on the characteristics of the histogram. For a given pixel, its saliency value is related to the histogram characteristics of the area where it is located. If the histogram of the area where the pixel is located is significantly different from the histogram of the entire image, the saliency value of the pixel is higher; conversely, if the histogram difference is small, the saliency value is lower. The saliency algorithm based on histogram contrast is simple, intuitive, and has high computational efficiency. In the embodiment of the present application, by combining saliency information, weak dynamic pixels that are uncertain whether they are moving can be classified as dynamic pixels, which can further improve the accuracy of motion and stillness judgments.

[0033] Step S206: performing noise processing on the dynamic pixels and the static pixels.

[0034] Noise processing is performed on the dynamic pixels and the static pixels, including: taking the original data corresponding to the dynamic pixels as a first time domain fusion result; performing time domain fusion on the original data corresponding to the static pixels and the time domain fusion result corresponding to the static pixels in the previous video frame to obtain a second time domain fusion result; and outputting the first time domain fusion result and the second time domain fusion result as the time domain fusion result of the current video frame.

[0035] The original data corresponding to the static pixel is temporally fused with the temporal fusion result corresponding to the static pixel in the previous video frame to obtain the second temporal fusion result, including: calculating the second temporal fusion result corresponding to the static pixel Among them, the fusion weight t is the time sequence of the current video frame, is the time domain fusion result of the previous video frame, is the noise estimate of the previous video frame, σ t is the preset noise estimation value of the current video frame, f t The original data of the current video frame.

[0036] The fusion weight is calculated by combining the noise level of the previous historical frame and the noise level of the current video frame, where f t is the pixel value of the current video frame. Different from the method in the related art which only considers the noise of the current video frame, the time domain fusion method in the embodiment of the present application ensures the fastest convergence speed, thereby enhancing the dynamic performance of the video frame in time.

[0037] Calculate the second time domain fusion result corresponding to the static pixel Afterwards, it also includes: the noise estimation value σ of the current video frame t Update to generate a second noise estimate

[0038] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0039] In this embodiment, a video frame noise processing device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0040] Figure 3 is a structural block diagram of a video frame noise processing device according to an embodiment of the present application, such as Figure 3 As shown, the device includes: a first setting module 10, a second setting module 20, and a noise processing module 30:

[0041] A first setting module 10 is used to classify each pixel of the current video frame as a dynamic pixel, a weak dynamic pixel or a static pixel;

[0042] A second setting module 20, configured to determine the weak dynamic pixel as a dynamic pixel or a static pixel according to pixel information and saliency information of a neighborhood of the weak dynamic pixel;

[0043] The noise processing module 30 is used to perform noise processing on the dynamic pixels and the static pixels.

[0044] Figure 4 is a structural block diagram of a video frame noise processing device according to another embodiment of the present application. Figure 4 As shown, the device includes Figure 3 In addition to all the modules shown, the first setting module further includes: a distinguishing unit 11, and the noise processing module further includes: an output unit 31:

[0045] The distinguishing unit 11 is used to subtract the pixel value of each target pixel in the previous frame from the pixel value of the target pixel in the current video frame, and compare the absolute value of the obtained difference with a first threshold and a second threshold, wherein when the absolute value of the difference is less than the first threshold, the target pixel is the static pixel; when the absolute value of the difference is greater than or equal to the first threshold and less than the second threshold, the target pixel is the weak dynamic pixel; when the absolute value of the difference is greater than the second threshold, the target pixel is the dynamic pixel.

[0046] The output unit 31 is used to use the original data corresponding to the dynamic pixels as the first time domain fusion result; perform time domain fusion on the original data corresponding to the static pixels and the time domain fusion result corresponding to the static pixels in the previous video frame to obtain a second time domain fusion result; and output the first time domain fusion result and the second time domain fusion result as the time domain fusion result of the current video frame.

[0047] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0048] To facilitate the understanding of the technical solution provided by the present application, a detailed explanation will be given below in combination with embodiments of specific scenarios.

[0049] Figure 5 is a structural diagram of a video frame noise processing system according to an embodiment of the present application, such as Figure 5 As shown in the figure, after the image is input, the image is firstly detected for saliency, and then the motion and static judgment and time domain fusion are performed. The motion and static judgment and time domain fusion require the use of the noise estimation module, and finally the denoised image is output.

[0050] Overall Figure 5 The input of the system shown includes: the denoising result of the previous frame The original data of the current frame f t , the noise estimate of the previous frame Current frame noise estimate σ t ; Output includes: current frame noise reduction result And its corresponding noise estimation result In the embodiment of the present application, only the noise reduction result of the current frame is stored. And the noise estimation results

[0051] The working principle of the saliency detection module in the video frame noise processing system of the embodiment of the present application is as follows: the original data f of the current frame is input t, output the saliency detection result S. This module is based on the original data f of the current frame t For saliency detection, the embodiment of the present application does not limit the specific algorithm of saliency detection, but only provides an optional algorithm: a saliency HC algorithm based on histogram contrast; specifically, the difference between the current pixel and other pixels in the image is calculated by formula (1). Given an input f t , the saliency size at the pixel point (x, y) is:

[0052]

[0053] After calculating the significance of the generated pixel point by formula (1), it can be accelerated by histogram statistics. After acceleration, the significance of the pixel point (x, y) can be calculated by formula (2):

[0054]

[0055] In formula (2), c i Indicates the color of the pixel (x, y), c j is a color value, n is the total number of colors, h i For color c i Probability of occurrence.

[0056] Figure 6 It is a schematic diagram of the effect of the saliency detection algorithm according to the embodiment of the present application. The HC saliency detection algorithm can achieve the effect of converting the original input into the color space LAB based on human eye perception.

[0057] The motion and stillness judgment module in the video frame noise processing system of the embodiment of the present application inputs the noise reduction result of the previous frame The original data of the current frame f t , the noise estimate of the previous frame Current frame noise estimate σ t , output the motion and stillness judgment result M of the current frame. The overall idea of ​​this module is similar to the secondary confirmation idea of ​​Canny edge detection. Figure 7 This is the overall decision diagram of the static and dynamic decision module. Figure 7 As shown, pixels are distinguished by means of dynamic and static decision initialization and dynamic and static decision optimization.

[0058] Specifically, first, Figure 8 The schematic diagram of motion-static judgment initialization shown in the figure uses a dual threshold method to distinguish each pixel as a dynamic pixel, a weak dynamic pixel or a static pixel, where the threshold setting is related to the noise level, and this part is named motion-static judgment initialization; then, the weak dynamic pixels are further distinguished as dynamic or static by combining the neighborhood information and the saliency information, and this part is named motion-static judgment optimization.

[0059] The first threshold th1 and the second threshold th2 are adjustable values ​​related to noise. A typical setting method is shown in formula (3) and formula (4):

[0060]

[0061]

[0062] Subtract the pixel value of each pixel in the previous frame from the pixel value in the current video frame, and compare the absolute value of the obtained difference with the first threshold and the second threshold, wherein when the absolute value of the difference is less than the first threshold, the pixel is a static pixel; when the absolute value of the difference is greater than or equal to the first threshold and less than the second threshold, the pixel is a weak dynamic pixel; when the absolute value of the difference is greater than the second threshold, the pixel is a dynamic pixel.

[0063] The method of optimizing the dynamic and static judgment is: if there is a dynamic pixel or a significant pixel in the neighborhood of a weak dynamic pixel, it is set to dynamic, otherwise it is static. The neighborhood size can be adjusted by parameters, such as 5 neighborhoods and 7 neighborhoods. The threshold of significance can also be adjusted by the parameter S_min, for example, when S(x,y)>S_min, the weak dynamic pixel is set to a dynamic pixel.

[0064] The input of the time domain fusion module in the video frame noise processing system of the embodiment of the present application is the noise reduction result of the previous frame The original data of the current frame f t , the noise estimation value of the previous frame Current frame noise estimate σ t , output the fusion result of the current frame like Fig. 9 The block diagram of the time domain fusion module shown in the figure calculates the fusion weights of the current frame noise and the previous frame noise to complete the time domain fusion.

[0065] Specifically, for dynamic regions, this module directly selects the result f of the current frame t As the fusion result; for static areas, this module combines the noise level of the previous historical frame and the noise level σ of the current frame t Calculate the fusion weights and use Wiener filtering to ensure that the result is a linear minimum mean square error estimate. The calculation method of the Wiener filter coefficient is shown in formula (5):

[0066]

[0067] The fusion result of the current frame f t The calculation formula is shown in formula (6):

[0068]

[0069] The above-mentioned time domain fusion method ensures the fastest convergence speed, thereby enhancing the dynamic performance.

[0070] The input of the noise estimation module in the video frame noise processing system of the embodiment of the present application is the noise estimation value of the previous frame. Current frame noise estimate σ t , time domain fusion weight w, output the noise estimation value after fusion of the current frame This module estimates and updates the noise level of the output result according to the fusion weight. The noise estimate after fusion of the current frame is The calculation method of is shown in formula (7):

[0071]

[0072] Through the embodiments of the present application, the following technical effects are achieved: the idea of ​​dual-threshold secondary confirmation improves the accuracy of motion and stillness judgment, and ultimately improves the video denoising effect in dynamic areas; the method of combining saliency information further improves the accuracy of motion and stillness judgment, and improves the final video denoising effect; the time domain fusion method of combining the previous and next frame noise estimation accelerates the convergence speed, making the dynamic performance of video denoising better; overall, the denoising effect of video frames is better, the dynamic response is faster, and the resource usage is lower.

[0073] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0074] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0075] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0076] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0077] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0078] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0079] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for processing video frame noise, characterized in that: include: Distinguish each pixel of the current video frame as a dynamic pixel, a weakly dynamic pixel or a static pixel; Determining the weak dynamic pixel as a dynamic pixel or a static pixel according to pixel information and saliency information of a neighborhood of the weak dynamic pixel; Noise processing is performed on the dynamic pixels and the static pixels.

2. The method according to claim 1, characterized in that Distinguishing each of the pixels of the current video frame as the dynamic pixel, the weak dynamic pixel or the static pixel, comprising: Subtract the pixel value of each target pixel in the previous frame from the pixel value of the target pixel in the current video frame, and compare the absolute value of the obtained difference with a first threshold and a second threshold, wherein when the absolute value of the difference is less than the first threshold, the target pixel is the static pixel; When the absolute value of the difference is greater than or equal to the first threshold and less than the second threshold, the target pixel is the weak dynamic pixel; When the absolute value of the difference is greater than the second threshold, the target pixel is the dynamic pixel.

3. The method according to claim 2, characterized in that Before comparing the absolute value of the pixel value of each target pixel in the current video frame minus the pixel value of the pixel in the previous frame with the first threshold and the second threshold, the method further includes: The first threshold and the second threshold are set according to the noise estimation value of the current video frame and the noise estimation value of the previous frame.

4. The method according to claim 1, characterized in that: Determining the weak dynamic pixel as the dynamic pixel or the static pixel according to the pixel information of the neighborhood of the weak dynamic pixel and the saliency information includes: Determine whether the dynamic pixel exists in the neighborhood of the weak dynamic pixel and whether the weak dynamic pixel is a significant pixel; When the dynamic pixel exists in the neighborhood of the weak dynamic pixel or the weak dynamic pixel is the significant pixel, determining the weak dynamic pixel as the dynamic pixel; When there is no dynamic pixel in the neighborhood of the weak dynamic pixel and the weak dynamic pixel is not a significant pixel, determining the weak dynamic pixel as the static pixel; Wherein, the neighborhood of the weak dynamic pixel is adjusted by a preset parameter.

5. The method according to claim 1, characterized in that: The dynamic pixels and the static pixels are subjected to noise processing, including: Taking the original data corresponding to the dynamic pixels as the first time domain fusion result; Performing time domain fusion on the original data corresponding to the static pixel and the time domain fusion result corresponding to the static pixel in the previous video frame to obtain a second time domain fusion result; The first time domain fusion result and the second time domain fusion result are output as the time domain fusion result of the current video frame.

6. The method according to claim 5, performing time domain fusion on the original data corresponding to the static pixel and the time domain fusion result corresponding to the static pixel in the previous video frame to obtain the second time domain fusion result, comprising: Calculate the second time domain fusion result corresponding to the static pixel Among them, the fusion weight t is the time sequence of the current video frame, is the time domain fusion result of the previous video frame, is the noise estimation value of the previous video frame, σ t is the preset noise estimation value of the current video frame, f t is the original data of the current video frame.

7. The method according to claim 6, characterized in that Calculate the second time domain fusion result corresponding to the static pixel After that, it also includes: The noise estimation value σ of the current video frame t Update to generate a second noise estimate 8. A video frame noise processing device, characterized in that: include: A first setting module, used for distinguishing each pixel of the current video frame as a dynamic pixel, a weak dynamic pixel or a static pixel; A second setting module, configured to determine the weak dynamic pixel as a dynamic pixel or a static pixel according to pixel information and saliency information of a neighborhood of the weak dynamic pixel; The noise processing module is used to perform noise processing on the dynamic pixels and the static pixels.

9. The device according to claim 8, characterized in that The first setting module also includes: A distinguishing unit is used to subtract the pixel value of each target pixel in the current video frame from the pixel value of the target pixel in the previous frame, and compare the absolute value of the obtained difference with a first threshold and a second threshold, wherein when the absolute value of the difference is less than the first threshold, the target pixel is the static pixel; when the absolute value of the difference is greater than or equal to the first threshold and less than the second threshold, the target pixel is the weak dynamic pixel; when the absolute value of the difference is greater than the second threshold, the target pixel is the dynamic pixel.

10. The device according to claim 8, characterized in that The noise processing module also includes: The output unit is used to use the original data corresponding to the dynamic pixels as the first time domain fusion result; perform time domain fusion on the original data corresponding to the static pixels and the time domain fusion result corresponding to the static pixels in the previous video frame to obtain a second time domain fusion result; and output the first time domain fusion result and the second time domain fusion result as the time domain fusion result of the current video frame.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 7 when executed by a processor.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 7 are implemented.

Images