Image noise reduction method

Abstract

The invention provides an image noise reduction method, which comprises the following steps of: performing down-sampling type decomposition on a current frame image in a target video to obtain a low-frequency component and a high-frequency component; performing compression transformation on the bit width of the low-frequency component to obtain a low-bit-width component; calculating a first motion probability image according to the low bit width component, and mapping a first time domain weighting parameter according to the first motion probability image; performing time domain filtering on the low bit width component according to the low bit width component and the first time domain weighting parameter to obtain a first filtering output component; and carrying out inverse transformation of the compression transformation on the first filtering output component to obtain the low-frequency component of the current frame image after noise reduction. According to the method, the image can be subjected to frequency band division, adaptive data compression is performed, the motion judgment accuracy can be improved, the operation complexity is reduced, and the implementation cost of image noise reduction is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an image noise reduction method. Background Art

[0002] Video noise reduction is an important part of image processing. Video noise reduction requires calculating the motion probability between frames and then mapping it into time-domain weighted parameters. In actual low-light scenes, the input signal-to-noise ratio is extremely low, and the accuracy of motion probability judgment is reduced. Traditional technology generally solves this problem by either adding pre-processing to pre-filter the input low-SNR signal, or increasing the computational complexity of motion probability judgment to enhance the robustness of the calculation results. Both of the above solutions require additional computing overhead and power consumption. The industry needs a new solution to reduce computing overhead and power consumption. Summary of the invention

[0003] The purpose of the present invention is to provide an image denoising method, which divides the image into frequency bands and performs adaptive data compression, thereby improving the accuracy of motion judgment, reducing computational complexity, and reducing the cost of implementing image denoising.

[0004] To achieve the above-mentioned purpose, the present invention provides an image denoising method, comprising the following steps: performing a downsampling type decomposition on a current frame image in a target video to obtain a low-frequency component and a high-frequency component; performing a compression transformation on the bit width of the low-frequency component to obtain a low-bit-width component; calculating a first motion probability image based on the low-bit-width component, and mapping a first time domain weighting parameter based on the first motion probability image; performing time domain filtering on the low-bit-width component based on the low-bit-width component and the first time domain weighting parameter to obtain a first filtered output component; and performing an inverse transformation of the compression transformation on the first filtered output component to obtain the low-frequency component of the current frame image after denoising.

[0005] Optionally, performing a downsampling type decomposition on the current frame image in the target video to obtain a low-frequency component and a high-frequency component specifically includes: calling a discrete wavelet transform algorithm to perform a downsampling type decomposition on the current frame image in the target video to obtain the low-frequency component and the high-frequency component.

[0006] Optionally, a predefined database is provided, wherein the predefined database includes a plurality of predefined image histogram data and a predefined dynamic range mapping curve having a corresponding relationship with the predefined image histogram data, and the compression transformation and the inverse transformation of the compression transformation specifically include: obtaining the first image histogram data of the current frame image; obtaining the dynamic range mapping curve matching the first image histogram data based on the first image histogram data and the predefined database; and performing the compression transformation and the inverse transformation of the compression transformation based on the dynamic range mapping curve matching the first image histogram data.

[0007] Optionally, the method of obtaining a dynamic range mapping curve that matches the first image histogram data based on the first image histogram data and the predefined database specifically includes: obtaining the second image histogram data of the previous frame image of the current frame image; calculating the similarity between the first image histogram data and the second histogram data, and if they are similar, the current frame image adopts the same dynamic range mapping curve as the previous frame image of the current frame image.

[0008] Optionally, the method of obtaining a dynamic range mapping curve matching the first image histogram data based on the first image histogram data and the predefined database specifically includes: calculating the similarity between the first histogram data and each of the predefined image histogram data in the predefined database, and selecting several of the predefined image histogram data that are most similar to the first histogram data; and performing a weighted average calculation based on several of the predefined image histogram data that are most similar to the first histogram data to obtain a dynamic range mapping curve matching the first image histogram data.

[0009] Optionally, the first motion probability image is calculated using a frame difference method.

[0010] Optionally, the low-bitwidth component is time-domain filtered according to the low-bitwidth component and the first time-domain weighted parameter to obtain a first filtered output component, specifically including: using the first filtered output component as the accumulated low-bitwidth component of the subsequent frame image of the current frame image; when processing to the subsequent frame image, redefining the subsequent frame image as the current frame image, and performing time-domain filtering on the low-bitwidth component according to the redefined low-bitwidth component of the current frame image, the first time-domain weighted parameter and the accumulated low-bitwidth component to obtain the first filtered output component of the redefined current frame image.

[0011] Optionally, it also includes obtaining a second filtered output component of the high-frequency component, performing upsampling type reconstruction based on the low-frequency component after denoising and the second filtered output component, and obtaining the current frame image after denoising; obtaining the second filtered output component specifically includes: judging whether the noise of the current frame image converges based on the first motion probability image; if not, performing time domain filtering on the high-frequency component according to the first time domain weighting parameter to obtain the second filtered output component; if converged, further judging whether there are isolated points in the first motion probability image; if there are no isolated points, performing time domain filtering on the high-frequency component according to the first time domain weighting parameter to obtain the second filtered output component; if there are isolated points, calculating the second motion probability image based on the high-frequency component, mapping the second time domain weighting parameter based on the second motion probability image, and performing time domain filtering on the high-frequency component according to the second time domain weighting parameter to obtain the second filtered output component.

[0012] Optionally, judging whether the noise of the current frame image has converged based on the first motion probability image specifically includes: counting the number of pixels in the first motion probability image whose motion probability values ​​are greater than a preset threshold; if the number of pixels whose motion probability is greater than the preset threshold accounts for a proportion in the first motion probability image that exceeds a preset proportion value, judging that the noise has not converged; otherwise, judging that the noise has converged.

[0013] Optionally, the determining whether there is an isolated point in the first motion probability image specifically includes: calculating a first difference value between a maximum value and a second maximum value of the motion probability value in a 3*3 neighborhood around any target pixel in the first motion probability image; calculating a second difference value between a minimum value and a second minimum value of the motion probability value in a 3*3 neighborhood around the target pixel; excluding the pixel points corresponding to the maximum value of the motion probability value and the pixel points corresponding to the minimum value of the motion probability value in the 3*3 neighborhood around the target pixel, and calculating the average value of the motion probability values ​​of the remaining 7 pixels; if at least one of the following conditions is met, the target pixel is determined to be an isolated point: the absolute value of the first difference value is greater than or equal to the average value; the absolute value of the second difference value is greater than or equal to the average value.

[0014] The present invention also provides an image denoising system, comprising: an image decomposition module, used to perform a downsampling type decomposition on a current frame image in a target video to obtain a low-frequency component and a high-frequency component; a data compression module, used to perform a compression transformation on the bit width of the low-frequency component to obtain a low-bit-width component; a motion probability and time domain parameter calculation module, used to calculate a first motion probability image based on the low-bit-width component, and map a first time domain weighting parameter based on the first motion probability image; a time domain filtering module, used to perform time domain filtering on the low-bit-width component based on the low-bit-width component and the first time domain weighting parameter to obtain a first filtered output component; and an image reconstruction module, used to perform an inverse transformation on the compression transformation of the first filtered output component to obtain the low-frequency component of the current frame image after denoising.

[0015] The present invention also provides an electronic device, comprising: a memory storing a computer program; a processor communicating with the memory and executing the image denoising method described in any one of the above claims when the computer program is called; and a display communicating with the processor and the memory and used to display a GUI interaction interface related to the image denoising method.

[0016] To achieve the above object, the present invention further provides a readable storage medium storing a computer program, which implements any of the above-mentioned image denoising methods when executed by a processor.

[0017] To achieve the above object, the present invention further provides a computer program product, which, when executed by a processor, implements any of the above-mentioned image denoising methods.

[0018] The image noise reduction method provided by the present invention has the following beneficial effects:

[0019] The present invention provides an image denoising method, comprising the following steps: performing downsampling type decomposition on the current frame image in the target video to obtain low-frequency components and high-frequency components; performing compression transformation on the bit width of the low-frequency component to obtain a low-bit width component; calculating a first motion probability image according to the low-bit width component, and mapping a first time domain weighting parameter according to the first motion probability image; performing time domain filtering on the low-bit width component according to the low-bit width component and the first time domain weighting parameter to obtain a first filtering output component; performing an inverse transformation of the compression transformation on the first filtering output component to obtain the low-frequency component after denoising of the current frame image. In this way, the present invention first performs downsampling type decomposition on the image, which can reduce the amount of data for subsequent core processing, and downsampling can improve the image signal-to-noise ratio. Then, the data is adaptively compressed in bit width, which can reduce the bit width of the data for subsequent core processing and reduce the computational complexity. At the same time, the adaptive bit width compression ensures the processing accuracy of the image signal concentration area. By filtering the signal in the frequency band, the signal-to-noise ratio of the low-frequency signal is better, and the time domain weight of the high-frequency component is replaced with its corresponding time domain weight as much as possible, which can simplify the calculation complexity and improve the accuracy of motion judgment. In summary, the present invention can divide the image into frequency bands and perform adaptive data compression, which can improve the accuracy of motion judgment, reduce the calculation complexity, and reduce the cost of image noise reduction.

[0020] The present invention also provides an image denoising system, which is used to execute any of the above-mentioned image denoising methods. Therefore, the image denoising system can divide the image into frequency bands and perform adaptive data compression, thereby improving the accuracy of motion judgment, reducing computational complexity, and reducing the cost of implementing image denoising.

[0021] The present invention also provides an electronic device, which is used to perform any of the above-mentioned image denoising methods. Therefore, the electronic device can divide the image into frequency bands and perform adaptive data compression, thereby improving the accuracy of motion judgment, reducing the computational complexity, and reducing the cost of implementing image denoising.

[0022] The present invention also provides a readable storage medium, which is used to execute any of the above-mentioned image denoising methods. Therefore, the readable storage medium can divide the image into frequency bands and perform adaptive data compression, thereby improving the accuracy of motion judgment, reducing the computational complexity, and reducing the implementation cost of image denoising.

[0023] The present invention also provides a computer program product, which is used to execute any of the above-mentioned image denoising methods. Therefore, the computer program product can divide the image into frequency bands and perform adaptive data compression, thereby improving the accuracy of motion judgment, reducing the computational complexity, and reducing the implementation cost of image denoising. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of a process flow of an image noise reduction method provided by an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of predefined image histogram data and a predefined dynamic range mapping curve having a corresponding relationship with the predefined image histogram data provided by an embodiment of the present invention;

[0026] Figure 3 A schematic diagram of a flow chart of an image noise reduction method provided by another embodiment of the present invention;

[0027] Figure 4 A schematic diagram of a block structure of an electronic device provided by an embodiment of the present invention;

[0028] The accompanying drawings are denoted as follows:

[0029] 101 - processor; 102 - communication interface; 103 - memory; 104 - communication bus; 105 - display. DETAILED DESCRIPTION

[0030] In order to make the purpose, advantages and features of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In addition, the structure shown in the drawings is often a part of the actual structure. In particular, the emphasis of each drawing is different, and sometimes different scales are used.

[0031] It should be understood that when an element or layer is referred to as "on...", "connected to" other elements or layers, it can be directly on other elements or layers, connected to other elements or layers, or there can be intervening elements or layers. On the contrary, when an element is referred to as "directly on...", "directly connected to" other elements or layers, there is no intervening element or layer. Although the terms first, second, third, etc. can be used to describe various elements, components, regions, layers and / or parts, these elements, components, regions, layers and / or parts should not be limited by these terms. These terms are only used to distinguish an element, component, region, layer or part from another element, component, region, layer or part. Therefore, without departing from the teachings of the present invention, the first element, component, region, layer or part discussed below can be expressed as a second element, component, region, layer or part. Spatial relationship terms such as "under...", "below", "below", "above...", "above", "above", etc., can be used here for the convenience of description to describe the relationship between an element or feature shown in the figure and other elements or features. It should be understood that, in addition to the orientation shown in the figure, the spatial relationship terms are intended to include different orientations of the devices in use and operation. For example, if the device in the accompanying drawings is turned over, then, the elements or features described as "under...", "below", "below" will be oriented to be "on" other elements or features. The device can be oriented in addition (rotated 90 degrees or other orientations) and the spatial descriptors used herein are interpreted accordingly. The purpose of the terms used herein is only to describe specific embodiments and is not intended to be a limitation of the present invention. When used herein, the singular forms of "one", "one" and "said / the" are also intended to include plural forms, unless the context clearly indicates another way. It should also be understood that the term "include" is used to determine the presence of features, steps, operations, elements and / or parts, but does not exclude the presence or addition of one or more other features, steps, operations, elements, parts and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0032] The purpose of the present invention is to provide an image noise reduction method, system, electronic device, storage medium and program product. Combined with augmented reality technology for intelligent news editing, the efficiency and quality of news editing can be significantly improved, and human errors can be reduced.

[0033] Please refer to Figure 1 , Figure 1 FIG. 1 is a flow chart of an image noise reduction method provided by an embodiment of the present invention. Figure 1 To achieve the above purpose, the present invention provides an image noise reduction method, comprising the following steps:

[0034] Perform downsampling type decomposition on the current frame image in the target video to obtain low-frequency components and high-frequency components;

[0035] Compressing and transforming the bit width of the low-frequency component to obtain a low-bit width component;

[0036] Calculating a first motion probability image according to the low bit width component, and mapping a first time domain weighting parameter according to the first motion probability image;

[0037] According to the low bit width component and the first time domain weighting parameter, the low bit width component is subjected to time domain filtering to obtain a first filtering output component;

[0038] An inverse transformation of the compression transformation is performed on the first filter output component to obtain the low-frequency component of the current frame image after noise reduction.

[0039] With such a configuration, the present invention first performs downsampling type decomposition on the image, which can reduce the amount of data for subsequent core processing, and downsampling can improve the image signal-to-noise ratio. Then, the data is adaptively compressed, which can reduce the data bit width for subsequent core processing and reduce the computational complexity. At the same time, the adaptive bit width compression ensures the processing accuracy of the image signal concentrated area. The signal is filtered in the frequency band, and the low-frequency signal has a better signal-to-noise ratio. Try to replace the time domain weight of the high-frequency component with its corresponding time domain weight, which can simplify the computational complexity, improve the motion judgment accuracy, and reduce the implementation cost of image denoising.

[0040] Specifically, the downsampling type decomposition is performed on the current frame image in the target video to obtain the low-frequency component and the high-frequency component, which specifically includes:

[0041] The discrete wavelet transform algorithm is called to perform downsampling type decomposition on the current frame image in the target video to obtain the low-frequency component and the high-frequency component. Discrete wavelet transform discretizes the scale and translation of the basic wavelet. In image processing, binary wavelets are often used as wavelet transform functions, that is, division is performed using integer powers of 2. It should be understood that other methods can also be used to implement downsampling type decomposition of the current frame image in the target video, which will not be repeated here.

[0042] Further, a predefined database is provided, the predefined database includes a plurality of predefined image histogram data and a predefined dynamic range mapping curve having a corresponding relationship with the predefined image histogram data, the compression transformation and the inverse transformation of the compression transformation specifically include:

[0043] Acquire first image histogram data of the current frame image;

[0044] Acquire a dynamic range mapping curve matching the first image histogram data according to the first image histogram data and the predefined database;

[0045] The compression transformation and the inverse transformation of the compression transformation are performed according to a dynamic range mapping curve matched with the first image histogram data.

[0046] Please refer to Figure 2 , Figure 2 A schematic diagram of predefined image histogram data and a predefined dynamic range mapping curve having a corresponding relationship with the predefined image histogram data provided by an embodiment of the present invention. Figure 2 In FIG. 1 , the three figures in the leftmost column are the predefined image histogram data, and the figure in the same row as the predefined image histogram data is a schematic diagram of a predefined dynamic range mapping curve having a corresponding relationship with the predefined image histogram data. Figure 2 In an embodiment of the present invention, each of the predefined image histogram data corresponds to two predefined dynamic range mapping curves, forming a pair of predefined dynamic range mapping curves, which are used to perform the compression transformation and the inverse transformation of the compression transformation respectively. Figure 2 In the embodiment of the present invention, three groups of predefined image histogram data and predefined dynamic range mapping curves corresponding to the predefined image histogram data are provided. However, in actual application, multiple groups may be provided, which will not be described in detail here.

[0047] When obtaining a dynamic range mapping curve that matches the first image histogram data based on the first image histogram data and the predefined database, it is necessary to determine the similarity between the current frame and the previous frame in the video dimension to determine whether the dynamic range mapping curve needs to be replaced, wherein the first frame will have an initial matching dynamic range mapping curve. Based on this, obtaining a dynamic range mapping curve that matches the first image histogram data based on the first image histogram data and the predefined database specifically includes: obtaining the second image histogram data of the previous frame image of the current frame image; calculating the similarity between the first image histogram data and the second histogram data, if similar, the current frame image adopts the same dynamic range mapping curve as the previous frame image of the current frame image.

[0048] For example, to calculate the current image histogram statistics. Assuming that the range is fixed and divided into 32 bins, we get the vector H c1×32 : Read the image histogram of the previous frame buffer: histogram data H p1×32 ; Calculate the similarity of the two histogram statistics. The similarity is determined by calculating the cosine distance Cosθ of the two sets of vectors. The specific relationship is as follows:

[0049]

[0050] If they are not similar, the dynamic range mapping curve needs to be transformed. The specific solution is as follows:

[0051] The method of obtaining a dynamic range mapping curve matching the first image histogram data based on the first image histogram data and the predefined database specifically includes: calculating the similarity between the first histogram data and each of the predefined image histogram data in the predefined database, and selecting several of the predefined image histogram data that are most similar to the first histogram data; and performing a weighted average calculation based on several of the predefined image histogram data that are most similar to the first histogram data to obtain a dynamic range mapping curve matching the first image histogram data.

[0052] For example, Figure 2 For the predefined database, the cosine similarity Cosθ between the current histogram and three typical histograms is calculated respectively. x It should be noted that in order to simplify the description, this embodiment only takes three typical histograms. In order to achieve more refined curve mapping in actual engineering, more types of histograms and corresponding basic curve pairs can be selected. x Select the two Cosθ that reflect the greatest similarity i ,Cosθ j , taking these two as weights, and the corresponding curve Crr i ,Crv j Perform linear interpolation and output the curve Crv c As the mapping curve of the current histogram. The specific relationship is as follows:

[0053]

[0054] It should be noted that the compression transformation may also be performed using a direct shift operation.

[0055] Specifically, the first motion probability image is calculated using a frame difference method. The frame difference method in image processing is a technology used for video analysis, mainly used to detect and extract changes between two frames in a video sequence, thereby achieving purposes such as motion detection and target tracking.

[0056] Furthermore, the performing time domain filtering on the low bit width component according to the low bit width component and the first time domain weighting parameter to obtain the first filtering output component specifically includes: using the first filtering output component as the accumulated low bit width component of the next frame image of the current frame image; when processing to the next frame image, redefining the next frame image as the current frame image, performing time domain filtering on the low bit width component according to the redefined low bit width component of the current frame image, the first time domain weighting parameter and the accumulated low bit width component, to obtain the redefined first filtering output component of the current frame image. For example, based on the low bit width component C LCalculate the first motion probability image MP L Then mapped into the first time domain weighting parameter TW L , mapping the first motion probability image into the time domain weighting parameter can be completed by using a predefined curve. L , the first filtering output component is F L , then the following relationship is satisfied:

[0057] F L =TW L C L +(1-TW L )Pre L

[0058] Furthermore, the high frequency components should also be processed. Figure 3 , Figure 3 FIG. 1 is a flow chart of an image noise reduction method provided by another embodiment of the present invention. Figure 3 As shown, it also includes obtaining a second filtered output component of the high-frequency component, performing upsampling type reconstruction according to the low-frequency component after noise reduction and the second filtered output component, and obtaining the current frame image after noise reduction; the obtaining the second filtered output component specifically includes:

[0059] Determining whether the noise of the current frame image converges according to the first motion probability image;

[0060] If it has not converged, performing time domain filtering on the high frequency component according to the first time domain weighted parameter to obtain the second filtering output component;

[0061] If converged, further determining whether there are isolated points in the first motion probability image;

[0062] If there is no isolated point, performing time domain filtering on the high frequency component according to the first time domain weighting parameter to obtain the second filtering output component;

[0063] If isolated points exist, a second motion probability image is calculated based on the high-frequency component, and a second time domain weighted parameter is mapped based on the second motion probability image. The high-frequency component is time-domain filtered based on the second time domain weighted parameter to obtain the second filtered output component.

[0064] The main logic of this setting is that if the noise of the current frame image converges and there are isolated points, the second motion probability image is further calculated for the high-frequency component; otherwise, the first time domain weighting parameter and the first motion probability image that are the same as the low-frequency component can be directly used.

[0065] When processing the high-frequency component, the processing logic is similar to that for the low-bit width component, and the second filtered output component is used as the accumulated high-frequency component of the next frame image of the current frame image; when processing the next frame image, the next frame image is redefined as the current frame image, and the high-frequency component is time-domain filtered according to the redefined high-frequency component of the current frame image, the second time-domain weighted parameter and the accumulated high-frequency component, to obtain the redefined second filtered output component of the current frame image. For example, based on the high-frequency component C H Calculate the first motion probability image MP H Then mapped into the second time domain weighting parameter TW H , mapping the second motion probability image into the time domain weighting parameter can be completed using a predefined curve. Let the accumulated high frequency component be Pre H , the first filtering output component is F H , then the following relationship is satisfied:

[0066] F H =TW H C H +(1-TW H )Pre H

[0067] Specifically, judging whether the noise of the current frame image is converged according to the first motion probability image specifically includes:

[0068] The number of pixels whose motion probability values ​​are greater than a preset threshold in the first motion probability image is counted; if the proportion of the number of pixels whose motion probability is greater than the preset threshold in the first motion probability image exceeds a preset proportion value, it is determined that the noise has not converged; otherwise, it is determined that the noise has converged.

[0069] Furthermore, the determining whether there are isolated points in the first motion probability image specifically includes:

[0070] Calculate a first difference value between a maximum value and a second maximum value of motion probability values ​​in a 3*3 neighborhood around any target pixel in the first motion probability image;

[0071] Calculate the second difference between the minimum value and the second minimum value of the motion probability value in the 3*3 neighborhood around the target pixel point; in the 3*3 neighborhood around the target pixel point, exclude the pixel point corresponding to the maximum value of the motion probability value and the pixel point corresponding to the minimum value of the motion probability value, and calculate the average value of the motion probability values ​​of the remaining 7 pixels;

[0072] If at least one of the following conditions is met, the target pixel is determined to be an isolated point:

[0073] The absolute value of the first difference value is greater than or equal to the average value;

[0074] An absolute value of the second difference value is greater than or equal to the average value.

[0075] The present invention also provides an image denoising system, comprising: an image decomposition module, used to perform a downsampling type decomposition on a current frame image in a target video to obtain a low-frequency component and a high-frequency component; a data compression module, used to perform a compression transformation on the bit width of the low-frequency component to obtain a low-bit-width component; a motion probability and time domain parameter calculation module, used to calculate a first motion probability image based on the low-bit-width component, and map a first time domain weighting parameter based on the first motion probability image; a time domain filtering module, used to perform time domain filtering on the low-bit-width component based on the low-bit-width component and the first time domain weighting parameter to obtain a first filtered output component; and an image reconstruction module, used to perform an inverse transformation on the compression transformation of the first filtered output component to obtain the low-frequency component of the current frame image after denoising.

[0076] Since the image denoising system is used to execute any of the above-mentioned image denoising methods, the image denoising system can divide the image into frequency bands and perform adaptive data compression, thereby improving the accuracy of motion judgment, reducing computational complexity, and reducing the cost of implementing image denoising.

[0077] Since the image noise reduction system provided by the present invention and the image noise reduction method described above belong to the same inventive concept, the electronic device provided by the present invention has all the advantages of the image noise reduction method described above, and therefore the beneficial effects of the electronic device provided by the present invention will not be described one by one here.

[0078] To achieve the above object, the present invention also provides an electronic device, please refer to Figure 4 , Figure 4 FIG. 1 is a block diagram of an electronic device provided by an embodiment of the present invention. Figure 4 As shown, the electronic device includes:

[0079] A memory 103 storing a computer program;

[0080] A processor 101 is connected to the memory for communication, and executes any one of the above-mentioned image noise reduction methods when calling the computer program;

[0081] The display 105 is communicatively connected with the processor and the memory, and is used to display a GUI interaction interface related to the image noise reduction method.

[0082] like Figure 4As shown, the electronic device also includes a communication interface 102 and a communication bus 104, wherein the processor 101, the communication interface 102, and the memory 103 communicate with each other through the communication bus 104. The communication bus 104 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 104 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface 102 is used for communication between the above-mentioned electronic device and other devices.

[0083] The processor 101 referred to in the present invention may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 101 is the control center of the electronic device, and various interfaces and lines are used to connect various parts of the entire electronic device.

[0084] The memory 103 may be used to store the computer program. The processor 101 implements various functions of the electronic device by running or executing the computer program stored in the memory 103 and calling the data stored in the memory 103 .

[0085] The memory 103 may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0086] Since the electronic device is used to execute any of the above-mentioned image denoising methods, the electronic device can divide the image into frequency bands and perform adaptive data compression, thereby improving the accuracy of motion judgment, reducing computational complexity, and reducing the cost of implementing image denoising.

[0087] Since the electronic device provided by the present invention and the image noise reduction method described above belong to the same inventive concept, the electronic device provided by the present invention has all the advantages of the image noise reduction method described above, and therefore the beneficial effects of the electronic device provided by the present invention will not be described one by one here.

[0088] To achieve the above object, the present invention provides a readable storage medium storing a computer program, which, when executed by a processor, implements the image denoising method as described in any one of the above. Since the readable storage medium is used to execute any one of the image denoising methods described in the above. Therefore, the readable storage medium can divide the image into frequency bands and perform adaptive data compression, which can improve the accuracy of motion judgment, reduce the computational complexity, and reduce the implementation cost of image denoising.

[0089] Since the readable storage medium provided by the present invention and the image denoising method described above belong to the same inventive concept, the readable storage medium provided by the present invention has all the advantages of the image denoising method described above, and therefore the beneficial effects of the readable storage medium provided by the present invention will not be described one by one here.

[0090] The readable storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer hard disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this article, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.

[0091] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program codes. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0092] The present invention also provides a computer program product, which implements any of the above-mentioned image noise reduction methods when executed by a processor.

[0093] The present invention also provides a computer program product, which is used to execute any of the above-mentioned image denoising methods. Therefore, the computer program product can divide the image into frequency bands and perform adaptive data compression, thereby improving the accuracy of motion judgment, reducing the computational complexity, and reducing the implementation cost of image denoising.

[0094] Since the computer program product provided by the present invention and the image denoising method described above belong to the same inventive concept, the computer program product provided by the present invention has all the advantages of the image denoising method described above, and therefore the beneficial effects of the computer program product provided by the present invention will not be described one by one here.

[0095] It should be noted that the various embodiments in this specification are described in a progressive manner, and any embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0096] It should also be noted that, although the present invention has been disclosed as a preferred embodiment, the above embodiment is not intended to limit the present invention. For any technician familiar with the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or modified into equivalent embodiments of equivalent changes. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of protection of the technical solution of the present invention.

[0097] It should also be understood that, unless otherwise specified or indicated, the terms "first", "second", "third", etc. in the specification are merely used to distinguish between the various components, elements, steps, etc. in the specification, and are not used to indicate the logical relationship or sequential relationship between the various components, elements, steps, etc.

[0098] It should also be recognized that the terms described herein are used only to describe specific embodiments and are not intended to limit the scope of the invention. It should be noted that the singular forms "a" and "an" used herein and in the appended claims include plural references unless the context clearly indicates otherwise. For example, a reference to "a step" or "a device" means a reference to one or more steps or devices, and may include secondary steps and secondary devices. All conjunctions used should be understood in the broadest sense. And, the word "or" should be understood to have the definition of a logical "or", rather than a logical "exclusive or", unless the context clearly indicates otherwise. In addition, the implementation of embodiments of the present invention may include performing selected tasks manually, automatically, or in combination.

Claims

1. An image denoising method, characterized in that: The following steps are involved: Perform downsampling type decomposition on the current frame image in the target video to obtain low-frequency components and high-frequency components; Compressing and transforming the bit width of the low-frequency component to obtain a low-bit width component; Calculating a first motion probability image according to the low bit width component, and mapping a first time domain weighting parameter according to the first motion probability image; According to the low bit width component and the first time domain weighting parameter, the low bit width component is subjected to time domain filtering to obtain a first filtering output component; An inverse transformation of the compression transformation is performed on the first filter output component to obtain the low-frequency component of the current frame image after noise reduction.

2. The image denoising method according to claim 1, wherein: The downsampling type decomposition of the current frame image in the target video to obtain the low-frequency component and the high-frequency component specifically includes: A discrete wavelet transform algorithm is called to perform downsampling type decomposition on the current frame image in the target video to obtain the low-frequency component and the high-frequency component.

3. The image denoising method according to claim 1, wherein: A predefined database is provided, wherein the predefined database includes a plurality of predefined image histogram data and a predefined dynamic range mapping curve having a corresponding relationship with the predefined image histogram data, and the compression transformation and the inverse transformation of the compression transformation specifically include: Acquire first image histogram data of the current frame image; Acquire a dynamic range mapping curve matching the first image histogram data according to the first image histogram data and the predefined database; The compression transformation and the inverse transformation of the compression transformation are performed according to a dynamic range mapping curve matched with the first image histogram data.

4. The image denoising method according to claim 3, wherein: The acquiring, according to the first image histogram data and the predefined database, a dynamic range mapping curve matching the first image histogram data specifically includes: Acquire second image histogram data of a frame image previous to the current frame image; The similarity between the first image histogram data and the second histogram data is calculated. If they are similar, the current frame image adopts the same dynamic range mapping curve as the previous frame image of the current frame image.

5. The image denoising method according to claim 3, wherein: The acquiring, according to the first image histogram data and the predefined database, a dynamic range mapping curve matching the first image histogram data specifically includes: Calculating the similarity between the first histogram data and each of the predefined image histogram data in the predefined database, and selecting a number of the predefined image histogram data that are most similar to the first histogram data; A weighted average calculation is performed based on a plurality of the predefined image histogram data that are most similar to the first histogram data to obtain a dynamic range mapping curve that matches the first image histogram data.

6. The image denoising method according to claim 1, wherein: The first motion probability image is calculated using a frame difference method.

7. The image denoising method according to claim 1, wherein: The step of performing time domain filtering on the low bit width component according to the low bit width component and the first time domain weighting parameter to obtain a first filtering output component specifically includes: Using the first filtering output component as the accumulated low-bit-width component of a frame image subsequent to the current frame image; When processing the subsequent frame image, the subsequent frame image is redefined as the current frame image, and the low-bitwidth component is time-domain filtered according to the redefined low-bitwidth component of the current frame image, the first time-domain weighted parameter and the accumulated low-bitwidth component to obtain the first filtered output component of the redefined current frame image.

8. The image denoising method according to claim 1, wherein: It also includes obtaining a second filtered output component of the high-frequency component, performing up-sampling type reconstruction according to the low-frequency component after denoising and the second filtered output component, and obtaining the current frame image after denoising; The obtaining of the second filtering output component specifically comprises: Determining whether the noise of the current frame image converges according to the first motion probability image; If it has not converged, performing time domain filtering on the high frequency component according to the first time domain weighted parameter to obtain the second filtering output component; If converged, further determining whether there are isolated points in the first motion probability image; If there is no isolated point, performing time domain filtering on the high frequency component according to the first time domain weighting parameter to obtain the second filtering output component; If isolated points exist, a second motion probability image is calculated based on the high-frequency component, and a second time domain weighted parameter is mapped based on the second motion probability image. The high-frequency component is time-domain filtered based on the second time domain weighted parameter to obtain the second filtered output component.

9. The image denoising method according to claim 8, wherein: The determining, according to the first motion probability image, whether the noise of the current frame image is converged specifically includes: The number of pixels whose motion probability values ​​are greater than a preset threshold in the first motion probability image is counted; if the proportion of the number of pixels whose motion probability is greater than the preset threshold in the first motion probability image exceeds a preset proportion value, it is determined that the noise has not converged; otherwise, it is determined that the noise has converged.

10. The image denoising method according to claim 8, wherein: The determining whether there is an isolated point in the first motion probability image specifically includes: Calculate a first difference value between a maximum value and a second maximum value of motion probability values ​​in a 3*3 neighborhood around any target pixel in the first motion probability image; Calculate a second difference value between the minimum value and the second minimum value of the motion probability value in a 3*3 neighborhood around the target pixel; In a 3*3 neighborhood around the target pixel, exclude the pixel corresponding to the maximum motion probability value and the pixel corresponding to the minimum motion probability value, and calculate the average motion probability value of the remaining 7 pixels; If at least one of the following conditions is met, the target pixel is determined to be an isolated point: The absolute value of the first difference value is greater than or equal to the average value; An absolute value of the second difference value is greater than or equal to the average value.

Images