Image denoising processing method, device and equipment and computer readable medium

Abstract

Embodiments of the present disclosure disclose image denoising processing method, device and equipment and computer readable medium. A specific implementation of the method comprises: obtaining a target coded image sequence corresponding to a target coded video; for a target coded image in the target coded image sequence, performing the following denoising processing: determining a first luminance deviation value sequence and a second luminance deviation value sequence; generating a first pixel deviation value sequence and a second pixel deviation value sequence; generating a third pixel deviation value sequence; performing weighted processing on each third pixel deviation value in the third pixel deviation value sequence and a value corresponding to a target coded pixel in the target coded image to generate a denoised target coded pixel, and obtaining a denoised target coded image. The implementation improves the quality of the denoised target coded image, thereby improving the quality of the denoised target coded video.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to image noise reduction processing methods, apparatus, devices, and computer-readable media. Background Technology

[0002] Image denoising methods are techniques for reducing noise in a target image. A common approach to denoising a target image is to input the target image into a pre-defined denoising convolutional neural network (DnCNN) to generate the denoised image. This pre-defined denoising convolutional neural network can be a denoising convolutional neural network (DnCNN).

[0003] However, the inventors discovered that when using the above method to denoise a target image, the following technical problems often arise:

[0004] First, the target image is denoised using a pre-defined denoising convolutional neural network, but it cannot denoise based on the previous frame of the target image, resulting in low quality of the denoised target image.

[0005] Second, when using a pre-set denoising convolutional neural network to denoise the target image, the edges in the target image are relatively blurry, making it impossible to determine whether there has been a change compared to the previous frame of the target image, resulting in low accuracy.

[0006] Third, because the dark areas in the target image are less bright after noise reduction, it is impossible to compensate for the dark areas, resulting in low image clarity.

[0007] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0008] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0009] Some embodiments of this disclosure provide image noise reduction processing methods, apparatuses, devices, and computer-readable media to solve one or more of the technical problems mentioned in the background section above.

[0010] In a first aspect, some embodiments of this disclosure provide an image denoising processing method, the method comprising: acquiring a target coded image sequence corresponding to a target coded video, wherein the target coded video is in chroma and luminance format, the chroma and luminance format including a first chroma format, a second chroma format, a first luminance format, and a second luminance format; for a target coded image in the target coded image sequence, in response to determining that the target coded image is not a first frame target coded image and that the target coded image is an undenoised target coded image, performing the following denoising processing: determining a first luminance deviation value sequence and a second luminance deviation value sequence based on the target coded image and the denoised previous frame target coded image corresponding to the target coded image; generating a first pixel deviation value sequence and a second pixel deviation value sequence based on the first luminance deviation value sequence and the second luminance deviation value sequence; generating a third pixel deviation value sequence based on the first pixel deviation value sequence and the second pixel deviation value sequence; weighting each third pixel deviation value in the third pixel deviation value sequence with the value corresponding to the target coded pixel in the target coded image to generate a denoised target coded pixel, thereby obtaining a denoised target coded image.

[0011] Secondly, some embodiments of this disclosure provide an image denoising processing apparatus, the apparatus comprising: an acquisition unit configured to acquire a target coded image sequence corresponding to a target coded video, wherein the target coded video is in chroma and luma format, the chroma and luma format including a first chroma format, a second chroma format, a first luma format, and a second luma format; and an execution unit configured to, for a target coded image in the target coded image sequence, in response to determining that the target coded image is not a first frame target coded image and that the target coded image is an undenoised target coded image, perform the following denoising processing: based on the target... The previous frame of the target encoded image, corresponding to the denoised target encoded image, is used to determine a first brightness deviation value sequence and a second brightness deviation value sequence. Based on the first brightness deviation value sequence and the second brightness deviation value sequence, a first pixel deviation value sequence and a second pixel deviation value sequence are generated. Based on the first pixel deviation value sequence and the second pixel deviation value sequence, a third pixel deviation value sequence is generated. Each third pixel deviation value in the third pixel deviation value sequence is weighted with the value corresponding to the target encoded pixel in the target encoded image to generate the denoised target encoded pixel, thus obtaining the denoised target encoded image.

[0012] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0013] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0014] The above-described embodiments of this disclosure have the following beneficial effects: the image denoising processing method of some embodiments of this disclosure can improve the quality of the denoised target encoded image, thereby improving the quality of the denoised target encoded video. Specifically, the reason for the low quality of the denoised target image is that: when using a preset denoising convolutional neural network to denoise the target image, it is impossible to denoise based on the previous frame of the target image, thus resulting in low quality of the denoised target image. Based on this, some embodiments of the image denoising processing method disclosed herein obtain a target coded image sequence corresponding to a target coded video, wherein the format of the target coded video is a chroma and luminance format, and the chroma and luminance format includes a first chroma format, a second chroma format, a first luminance format, and a second luminance format; for the target coded image in the target coded image sequence, in response to determining that the target coded image is not the first frame target coded image and that the target coded image is an undenoised target coded image, the following denoising processing is performed: based on the target coded image and the denoised previous frame target coded image corresponding to the target coded image, a first luminance deviation value sequence and a second luminance deviation value sequence are determined; wherein the first luminance deviation value in the first luminance deviation value sequence may be the deviation value between the target coded pixel of the first luminance format in the target coded image and the denoised target coded pixel of the first luminance format in the denoised previous frame target coded image. The second brightness deviation value in the aforementioned second brightness deviation value sequence can be the deviation value between the target coded pixel in the second brightness format in the aforementioned target coded image and the denoised target coded pixel in the second brightness format of the aforementioned denoised previous frame target coded image. The first brightness deviation value and the second brightness deviation value are determined using the aforementioned target coded image and the aforementioned denoised previous frame target coded image, resulting in more accurate first and second brightness deviation values. Based on the aforementioned first brightness deviation value sequence and the aforementioned second brightness deviation value sequence, a first pixel deviation value sequence and a second pixel deviation value sequence are generated. The first pixel deviation value in the aforementioned first pixel deviation value sequence can be the deviation value between the target coded pixel in the first chroma format in the aforementioned target coded image and the denoised target coded pixel in the first chroma format of the aforementioned denoised previous frame target coded image, and the deviation value of the first brightness deviation value in the aforementioned first brightness deviation value sequence. The second pixel deviation value in the above second pixel deviation value sequence may be the deviation value of the target coded pixel in the second chroma format in the above target coded image, the deviation value of the target coded pixel in the second chroma format of the above denoised previous frame target coded image, and the deviation value of the second brightness deviation value in the above second brightness deviation value sequence.Based on the first pixel deviation value sequence and the second pixel deviation value sequence, a third pixel deviation value sequence is generated. The third pixel deviation value can be obtained by stitching together the first and second pixel deviation values ​​after exposure compensation. Each third pixel deviation value in the third pixel deviation value sequence is weighted with the corresponding value of the target coded pixel in the target coded image to generate a denoised target coded pixel, resulting in a denoised target coded image. By visualizing the data, a preset weighted value corresponding to the third pixel deviation value is obtained, and weighted with the corresponding target coded pixel to obtain the denoised target coded pixel. This improves the quality of the denoised target coded image, thereby improving the quality of the denoised target coded video. Attached Figure Description

[0015] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0016] Figure 1 This is a flowchart of some embodiments of the image noise reduction processing method according to the present disclosure;

[0017] Figure 2 These are schematic diagrams illustrating the structure of some embodiments of the image noise reduction processing apparatus according to the present disclosure;

[0018] Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0019] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0020] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0021] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0022] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0023] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0024] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] Figure 1 A flow 100 of some embodiments of the image denoising processing method according to the present disclosure is shown. The flow 100 of the image denoising processing method includes the following steps:

[0026] Step 101: Obtain the target encoded image sequence corresponding to the target encoded video.

[0027] In some embodiments, the entity executing the image denoising processing method (e.g., an electronic device) can acquire the target coded image sequence corresponding to the target coded video via a wired connection or a wireless connection. The target coded video is in chroma and luma format. The chroma and luma format can be YUV format. The chroma and luma format includes a first chroma format, a second chroma format, a first luma format, and a second luma format. The first chroma format can be the U format in YUV format. The second chroma format can be the V format in YUV format. The first luma format can be the Y format corresponding to the U format in YUV format. The second luma format can be the Y format corresponding to the V format in YUV format. The target coded video can be the video to be denoised. The target coded image sequence can be an image sequence of the target coded images in the target coded video ordered chronologically.

[0028] Step 102: For the target coded image in the target coded image sequence, in response to determining that the target coded image is not the first frame target coded image and that the target coded image is an undenoised target coded image, wherein the aforementioned first frame target coded image can be the earliest corresponding target coded image in the target coded image sequence, perform the following denoising processing:

[0029] Step 1021: Determine the first brightness deviation value sequence and the second brightness deviation value sequence based on the target coded image and the target coded image corresponding to the previous frame after noise reduction.

[0030] In some embodiments, the execution entity can determine a first brightness deviation value sequence and a second brightness deviation value sequence based on the target coded image and the corresponding denoised previous frame target coded image. The first brightness deviation value can be the standard deviation between the target coded image and the corresponding denoised previous frame target coded image. The second brightness deviation value can be the standard deviation between the target coded image and the corresponding denoised previous frame target coded image.

[0031] In practice, the first luminance deviation value in the above first luminance deviation value sequence can be obtained through the following steps:

[0032] The first step is to determine the target encoded pixels of the first brightness format in the target encoded image as the first brightness deviation value.

[0033] The second step is to determine the denoised target encoded pixels of the previous frame in the first brightness format in the above-mentioned denoised target encoded image as the second brightness deviation value.

[0034] The third step is to determine the first brightness deviation value as the deviation between the first brightness deviation value and the second brightness deviation value.

[0035] In practice, the second brightness deviation value in the above second brightness deviation value sequence can be obtained through the following steps:

[0036] The first step is to determine the target encoded pixels in the second brightness format of the target encoded image as the third brightness deviation value.

[0037] The second step is to determine the denoised target encoded pixels of the previous frame in the second brightness format in the above-mentioned denoised target encoded image as the fourth brightness deviation value.

[0038] The third step is to determine the deviation between the third and fourth brightness deviation values ​​as the second brightness deviation value.

[0039] In some optional implementations of certain embodiments, determining the first luminance deviation value sequence and the second luminance deviation value sequence based on the target coded image and the corresponding denoised previous frame target coded image may include the following steps:

[0040] The first step is to determine the first deviation value as the difference between the average pixel value in the first pixel block and the average pixel value in the second pixel block. Here, the aforementioned denoised previous frame target coded image includes the first pixel block, and the aforementioned target coded image includes the second pixel block. The first pixel block may be composed of multiple pixels of the first chroma format in the aforementioned denoised previous frame target coded image, and the aforementioned second pixel block may be composed of multiple pixels of the first chroma format in the aforementioned target coded image.

[0041] The second step is to determine the second deviation value as the difference between the average pixel value in the third pixel block and the average pixel value in the fourth pixel block. Here, the aforementioned denoised previous frame target coded image includes a third pixel block, and the aforementioned target coded image includes a fourth pixel block. The third pixel block is composed of multiple pixels in the second chroma format from the aforementioned denoised previous frame target coded image, and the aforementioned fourth pixel block is composed of multiple pixels in the second chroma format from the aforementioned target coded image.

[0042] The third step is to determine the color deviation value by summing the first absolute value and the second absolute value. Here, the first absolute value is the absolute value of the first deviation value, and the second absolute value is the absolute value of the second deviation value.

[0043] The fourth step involves dividing the fifth pixel block into an average number of first sub-pixel blocks, which are then used as the first sub-pixel block sequence. The fifth pixel block is part of the previously encoded target image after noise reduction, and it is composed of multiple pixels of the first luminance format from the previously encoded target image after noise reduction.

[0044] The fifth step involves dividing the sixth pixel block into an average number of second sub-pixel blocks, which are then used as a sequence of second sub-pixel blocks. The target encoded image includes a sixth pixel block, which is composed of multiple pixels of the first brightness format in the target encoded image.

[0045] The sixth step involves dividing the seventh pixel block into an average number of third sub-pixel blocks, which are then used as a sequence of third sub-pixel blocks. The seventh pixel block is part of the previously encoded target image after noise reduction, and it is composed of multiple pixels of the second luminance format from the previously encoded target image after noise reduction.

[0046] Step 7: Divide the eighth pixel block into an average number of fourth sub-pixel blocks, which are then used as a sequence of fourth sub-pixel blocks. The target encoded image includes an eighth pixel block, which is composed of multiple pixels of the second luminance format in the target encoded image.

[0047] Step 8: Perform a first preset average value processing on each first sub-pixel block in the first sub-pixel block sequence and its corresponding second sub-pixel block in the second sub-pixel block sequence to generate a first brightness deviation value, thus obtaining a first brightness deviation value sequence. There is a one-to-one correspondence between the first sub-pixel blocks in the first sub-pixel block sequence and the second sub-pixel blocks in the second sub-pixel block sequence.

[0048] In practice, the above-mentioned method of performing a first preset average value processing on each first sub-pixel block in the first sub-pixel block sequence and the corresponding second sub-pixel block in the second sub-pixel block sequence to generate a first brightness deviation value may include the following steps:

[0049] The first sub-step involves determining the difference between the average pixel value in the first sub-pixel block and the average pixel value in the second sub-pixel block as the first average brightness deviation value.

[0050] The second sub-step is to determine the absolute value of the first average brightness deviation value as the first brightness deviation value.

[0051] Step 9: Perform a second preset average value processing on each third sub-pixel block in the third sub-pixel block sequence and its corresponding fourth sub-pixel block in the fourth sub-pixel block sequence to generate a second brightness deviation value, thus obtaining a second brightness deviation value sequence. There is a one-to-one correspondence between the third sub-pixel blocks in the third sub-pixel block sequence and the fourth sub-pixel blocks in the fourth sub-pixel block sequence.

[0052] In practice, the above-mentioned method of performing a second preset average value processing on each third sub-pixel block in the third sub-pixel block sequence and the corresponding fourth sub-pixel block in the fourth sub-pixel block sequence to generate a second brightness deviation value may include the following steps:

[0053] The first sub-step is to determine the second average brightness deviation value as the difference between the average pixel value in the third sub-pixel block and the average pixel value in the fourth sub-pixel block.

[0054] The second sub-step involves determining the absolute value of the aforementioned second average brightness deviation value as the second brightness deviation value.

[0055] Step 1022: Generate a first pixel deviation value sequence and a second pixel deviation value sequence based on the first brightness deviation value sequence and the second brightness deviation value sequence.

[0056] In some embodiments, a first pixel deviation value sequence and a second pixel deviation value sequence are generated based on a first brightness deviation value sequence and a second brightness deviation value sequence.

[0057] In practice, the first pixel deviation value in the above first pixel deviation value sequence can be obtained through the following steps:

[0058] The first step is to determine the target encoded pixel of the first brightness format in the target encoded image as the first pixel deviation value.

[0059] The second step is to determine the denoised target encoded pixel of the previous frame in the first brightness format in the above-mentioned denoised target encoded image as the second pixel deviation value.

[0060] The third step is to determine the deviation between the first pixel deviation value and the second pixel deviation value as the first brightness deviation value.

[0061] In practice, the second pixel deviation value in the above second pixel deviation value sequence can be obtained through the following steps:

[0062] The first step is to determine the target encoded pixel in the second brightness format of the target encoded image as the third pixel deviation value.

[0063] The second step is to determine the denoised target encoded pixel of the previous frame in the second brightness format in the above-mentioned denoised target encoded image as the fourth pixel deviation value.

[0064] The third step is to determine the deviation between the third pixel deviation value and the fourth pixel deviation value as the second pixel deviation value.

[0065] In some optional implementations of certain embodiments, generating the first pixel deviation value sequence and the second pixel deviation value sequence based on the first brightness deviation value sequence and the second brightness deviation value sequence may include the following steps:

[0066] The first step is to determine the average value of the first brightness deviation value sequence as the first overall deviation value, and to determine the average value of the second brightness deviation value sequence as the second overall deviation value.

[0067] The second step is to determine the absolute value of the difference between the value corresponding to each pixel in the fifth pixel block and the first overall deviation value mentioned above as the third overall deviation value, thus obtaining the third overall deviation value sequence.

[0068] The third step is to determine the absolute value of the difference between the value corresponding to each pixel in the seventh pixel block and the second overall deviation value mentioned above as the fourth overall deviation value, thus obtaining the fourth overall deviation value sequence.

[0069] The fourth step is to determine the fifth overall deviation value by multiplying the first overall deviation value by the first preset threshold. The first preset threshold can be a single numerical value. For example, the first preset threshold can be any value in the range [0,1].

[0070] The fifth step is to multiply the second overall deviation value and the first preset threshold value to determine the sixth overall deviation value.

[0071] Step 6: Determine the difference between the first preset value and the aforementioned first preset threshold as the first adjustment threshold. The aforementioned first preset value can be a single numerical value. The aforementioned first preset value can be the numerical value "1".

[0072] Step 7: The product of each first brightness deviation value in the first brightness deviation value sequence and the first adjustment threshold is determined as the seventh overall deviation value, thus obtaining the seventh overall deviation value sequence.

[0073] Step 8: The product of each second brightness deviation value in the second brightness deviation value sequence and the first adjustment threshold is determined as the eighth overall deviation value, thus obtaining the eighth overall deviation value sequence.

[0074] The ninth step is to determine the first comprehensive deviation value by summing the fifth overall deviation value and each of the seventh overall deviation values ​​in the above-mentioned seventh overall deviation value sequence, thus obtaining the first comprehensive deviation value sequence.

[0075] Step 10: The sum of the sixth overall deviation value and each of the eighth overall deviation values ​​in the above-mentioned eighth overall deviation value sequence is determined as the second comprehensive deviation value, thus obtaining the second comprehensive deviation value sequence.

[0076] The eleventh step is to determine the ninth overall deviation value by multiplying the second preset threshold by the chromaticity deviation value. The second preset threshold can be a single numerical value. For example, the first preset threshold can be any value in the range [0, 4].

[0077] Step 12: The product of the third preset threshold and each of the first comprehensive deviation values ​​in the aforementioned first comprehensive deviation value sequence is used to determine the third comprehensive deviation value, thus obtaining the third comprehensive deviation value sequence. The aforementioned third preset threshold can be a numerical value. For example, the aforementioned first preset threshold can be any value in [0, 4].

[0078] Step 13: The product of the third preset threshold and each second comprehensive deviation value in the second comprehensive deviation value sequence is determined as the fourth comprehensive deviation value, thus obtaining the fourth comprehensive deviation value sequence.

[0079] Step fourteen: The product of the fourth preset threshold and each of the third overall deviation values ​​in the above-mentioned third overall deviation value sequence is used to determine the tenth overall deviation value, thus obtaining the tenth overall deviation value sequence. Here, the fourth preset threshold can be a numerical value. For example, the first preset threshold can be any value in [0, 4].

[0080] Step 15: The product of the above-mentioned fourth preset threshold and each of the fourth overall deviation values ​​in the above-mentioned fourth overall deviation value sequence is determined as the eleventh overall deviation value, thus obtaining the eleventh overall deviation value sequence.

[0081] Step sixteen: Based on the above-mentioned ninth overall deviation value, the above-mentioned third comprehensive deviation value sequence, and the above-mentioned tenth overall deviation value sequence, determine the first pixel deviation value sequence.

[0082] In practice, the first positive third comprehensive deviation value to the fourth positive third comprehensive deviation value in the aforementioned third comprehensive deviation value sequence are defined as the first correspondence sequence. Furthermore, this first correspondence sequence has a corresponding relationship with the second positive tenth overall deviation value in the aforementioned tenth overall deviation value sequence.

[0083] In practice, the fifth to eighth positive third comprehensive deviation values ​​in the aforementioned third comprehensive deviation value sequence are defined as the second correspondence sequence. This second correspondence sequence corresponds to the second positive tenth overall deviation value in the aforementioned tenth overall deviation value sequence.

[0084] In practice, for each of the third comprehensive deviation values ​​in the above sequence, the following steps are performed:

[0085] The product of the aforementioned third comprehensive deviation value and the tenth overall deviation corresponding to the aforementioned third comprehensive deviation value, and the product of the aforementioned ninth overall deviation value, are determined as the first pixel deviation value, thus obtaining the first pixel deviation value sequence.

[0086] Step seventeen: Based on the aforementioned ninth overall deviation value, the aforementioned fourth comprehensive deviation value sequence, and the aforementioned eleventh overall deviation value sequence, determine the second pixel deviation value sequence. Specifically, the first to fourth positive fourth comprehensive deviation values ​​in the aforementioned fourth comprehensive deviation value sequence correspond to the first positive eleventh overall deviation value in the aforementioned eleventh overall deviation value sequence. Similarly, the fifth to eighth positive fourth comprehensive deviation values ​​in the aforementioned fourth comprehensive deviation value sequence correspond to the second positive eleventh overall deviation value in the aforementioned eleventh overall deviation value sequence. And so on.

[0087] In practice, for each of the fourth comprehensive deviation values ​​in the above sequence, the following steps are performed:

[0088] The product of the fourth overall deviation value and the eleventh overall deviation corresponding to the fourth overall deviation value, and the product of the ninth overall deviation value, are determined as the second pixel deviation value, thus obtaining the second pixel deviation value sequence.

[0089] The above-mentioned content, as an inventive point of this disclosure, solves the second technical problem mentioned in the background art: "When using a preset denoising convolutional neural network to denoise a target image, the edge parts of the target image are relatively blurry, making it impossible to determine whether there has been a change compared to the previous frame, resulting in low accuracy." The factors that lead to low accuracy due to the inability to determine whether there has been a change compared to the previous frame are often as follows: When using a preset denoising convolutional neural network to denoise a target image, the edge parts of the target image are relatively blurry, making it impossible to determine whether there has been a change compared to the previous frame, resulting in low accuracy. If these factors are resolved, a high accuracy for dangerous values ​​can be achieved. To achieve this effect, firstly, the average value of the first brightness deviation value sequence is determined as the first overall deviation value, and the average value of the second brightness deviation value sequence is determined as the second overall deviation value. Secondly, the absolute value of the difference between the value corresponding to each pixel in the fifth pixel block and the first overall deviation value is determined as the third overall deviation value, resulting in a third overall deviation value sequence. Third, the absolute value of the difference between the value corresponding to each pixel in the seventh pixel block and the aforementioned second overall deviation value is determined as the fourth overall deviation value, resulting in a fourth overall deviation value sequence. Fourth, the product of the aforementioned first overall deviation value and the first preset threshold is determined as the fifth overall deviation value. The aforementioned first preset threshold can be a single numerical value. For example, the aforementioned first preset threshold can be any value in [0,1]. Fifth, the product of the aforementioned second overall deviation value and the aforementioned first preset threshold is determined as the sixth overall deviation value. Sixth, the difference between the first preset value and the aforementioned first preset threshold is determined as the first adjustment threshold. The aforementioned first preset value can be a single numerical value. The aforementioned first preset value can be the value "1". Seventh, the product of each first brightness deviation value in the aforementioned first brightness deviation value sequence and the aforementioned first adjustment threshold is determined as the seventh overall deviation value, resulting in a seventh overall deviation value sequence. Eighth, the product of each second brightness deviation value in the aforementioned second brightness deviation value sequence and the aforementioned first adjustment threshold is determined as the eighth overall deviation value, resulting in an eighth overall deviation value sequence. Ninth, the sum of the fifth overall deviation value and each of the seventh overall deviation values ​​in the seventh overall deviation value sequence is determined as the first comprehensive deviation value, resulting in the first comprehensive deviation value sequence. Tenth, the sum of the sixth overall deviation value and each of the eighth overall deviation values ​​in the eighth overall deviation value sequence is determined as the second comprehensive deviation value, resulting in the second comprehensive deviation value sequence. The first and second comprehensive deviation values ​​may correspond to the noise level in the image. When the image noise is relatively weak, the first and second comprehensive deviation values ​​will be smaller, increasing sensitivity to local changes and avoiding excessive blockiness.When the image has strong noise, the first and second comprehensive deviation values ​​will be larger to suppress the influence of noise on the estimation of changes and to avoid image blurring. Eleventh, the product of the second preset threshold and the chromaticity deviation value is determined as the ninth overall deviation value. The second preset threshold can be a single numerical value. For example, the first preset threshold can be any value in [0,4]. Twelfth, the product of the third preset threshold and each first comprehensive deviation value in the first comprehensive deviation value sequence is determined as the third comprehensive deviation value, resulting in a third comprehensive deviation value sequence. The third preset threshold can be a single numerical value. For example, the first preset threshold can be any value in [0,4]. Thirteenth, the product of the third preset threshold and each second comprehensive deviation value in the second comprehensive deviation value sequence is determined as the fourth comprehensive deviation value, resulting in a fourth comprehensive deviation value sequence. Fourteenth, the product of the fourth preset threshold and each third overall deviation value in the third overall deviation value sequence is determined as the tenth overall deviation value, resulting in a tenth overall deviation value sequence. The fourth preset threshold can be a single numerical value. For example, the first preset threshold can be any value in [0,4]. Fifteenth, the product of the aforementioned fourth preset threshold and each of the fourth overall deviation values ​​in the aforementioned fourth overall deviation value sequence is determined as the eleventh overall deviation value, resulting in the eleventh overall deviation value sequence. Sixteenth, based on the aforementioned ninth overall deviation value, the aforementioned third comprehensive deviation value sequence, and the aforementioned tenth overall deviation value sequence, the first pixel deviation value sequence is determined. Specifically, the first to fourth positive third comprehensive deviation values ​​in the aforementioned third comprehensive deviation value sequence correspond to the first positive tenth overall deviation value in the aforementioned tenth overall deviation value sequence. The fifth to eighth positive third comprehensive deviation values ​​in the aforementioned third comprehensive deviation value sequence correspond to the second positive tenth overall deviation value in the aforementioned tenth overall deviation value sequence. And so on. In practice, for each third comprehensive deviation value in the aforementioned third comprehensive deviation value sequence, the following steps are performed: the product of the aforementioned third comprehensive deviation value and the corresponding tenth overall deviation, and the product of the aforementioned ninth overall deviation value, is determined as the first pixel deviation value, resulting in the first pixel deviation value sequence. Seventeenth, based on the aforementioned ninth overall deviation value, the aforementioned fourth comprehensive deviation value sequence, and the aforementioned eleventh overall deviation value sequence, determine the second pixel deviation value sequence. Specifically, the first to fourth positive fourth comprehensive deviation values ​​in the aforementioned fourth comprehensive deviation value sequence correspond to the first positive eleventh overall deviation value in the aforementioned eleventh overall deviation value sequence. Similarly, the fifth to eighth positive fourth comprehensive deviation values ​​in the aforementioned fourth comprehensive deviation value sequence correspond to the second positive eleventh overall deviation value in the aforementioned eleventh overall deviation value sequence. And so on.In practice, for each of the fourth comprehensive deviation values ​​in the aforementioned fourth comprehensive deviation value sequence, the following steps are performed: the product of the aforementioned fourth comprehensive deviation value and the corresponding eleventh overall deviation, and the product of the aforementioned ninth overall deviation value, are used to determine the second pixel deviation value, thus obtaining the second pixel deviation value sequence. The aforementioned first pixel deviation value sequence can be a deviation value sequence that integrates the aforementioned nine overall deviation values, the third comprehensive deviation value sequence, and the aforementioned tenth overall deviation value sequence. The aforementioned second pixel deviation value sequence can also be a deviation value sequence that integrates the aforementioned nine overall deviation values, the fourth comprehensive deviation value sequence, and the aforementioned eleventh overall deviation value sequence. The magnitudes of the values ​​in the aforementioned first pixel deviation value sequence and the aforementioned second pixel deviation value sequence can be determined based on noise. When the noise is relatively weak, the first pixel deviation value and the second pixel deviation value can be smaller to increase the sensitivity to differences in local changes and avoid repeatedly determining whether the target encoded image of the current frame has changed compared with the target encoded image of the previous frame after noise reduction. When the noise is relatively strong, the first pixel deviation value and the second pixel deviation value can avoid blurring the image and increase the sensitivity to local changes, making it easier to determine whether the target encoded image of the current frame has changed compared with the target encoded image of the previous frame after noise reduction.

[0090] Step 1023: Generate a third pixel deviation value sequence based on the first pixel deviation value sequence and the second pixel deviation value sequence.

[0091] In some embodiments, a third pixel deviation value sequence is generated based on the first pixel deviation value sequence and the second pixel deviation value sequence. The third pixel deviation value can be obtained by stitching together the first pixel deviation value and the second pixel deviation value after exposure compensation.

[0092] In some optional implementations of certain embodiments, generating a third pixel deviation value sequence based on the first pixel deviation value sequence and the second pixel deviation value sequence may include the following steps:

[0093] The first step involves performing compensation processing on the first overall deviation value and each first pixel deviation value in the aforementioned first pixel deviation value sequence to obtain a first compensated deviation value sequence. Here, the compensation processing can characterize the process of compensating for each first pixel deviation value in the first overall deviation value and the aforementioned first pixel deviation value sequence. For example, the compensation processing can be exposure compensation. The first compensated deviation value in the aforementioned first compensated deviation value sequence can be the deviation value after performing compensation processing on each first pixel deviation value in the aforementioned first overall deviation value and the aforementioned first pixel deviation value sequence.

[0094] The second step involves performing compensation processing on the second overall deviation value and each second pixel deviation value in the aforementioned second pixel deviation value sequence to obtain a second compensated deviation value sequence. This compensation processing can characterize the compensation process performed on the first overall deviation value and each first pixel deviation value in the aforementioned first pixel deviation value sequence. For example, this compensation processing can be exposure compensation. The second compensated deviation value in the aforementioned second compensated deviation value sequence can be the deviation value after performing compensation processing on each second pixel deviation value in the aforementioned second overall deviation value and the aforementioned second pixel deviation value sequence. The method for performing compensation processing on the second overall deviation value and each second pixel deviation value in the aforementioned second pixel deviation value sequence is consistent with the method for performing compensation processing on the first overall deviation value and each first pixel deviation value in the aforementioned first pixel deviation value sequence, and will not be repeated here.

[0095] The third step involves performing a first transformation process on the aforementioned first compensation deviation value sequence to obtain a first transformation value and a first transformation deviation value. This first transformation process can be determined by processing the image change information of the previous frame of the denoised target coded image, specifically whether the content of the previous frame of the denoised target coded image has changed compared to the previous frame of the denoised target coded image. The image change information of the previous frame of the denoised target coded image can be information about content changes compared to the previous frame of the denoised target coded image.

[0096] The fourth step involves performing a first transformation process on the aforementioned second compensation deviation value sequence to obtain a second transformation value and a second transformation deviation value. The method for performing the first transformation process on the aforementioned second compensation deviation value sequence is the same as the method for performing the first transformation process on the aforementioned first compensation deviation value sequence, and will not be repeated here.

[0097] The fifth step involves performing a second change processing on the first and second change values ​​to obtain the image change information of the target encoded image.

[0098] The sixth step is to concatenate the first and second variation deviation value sequences to obtain the third pixel deviation value sequence.

[0099] In some optional implementations of certain embodiments, the above-described first transformation process on the first compensation deviation value sequence to obtain a first change value and a first change deviation value may include the following steps:

[0100] The first step is to determine the aforementioned second preset value as the preset change value. The aforementioned second preset value can be a single numerical value. For example, the aforementioned second preset value can be the value "0".

[0101] The second step involves performing the following determination steps for each first compensation deviation value in the aforementioned first compensation deviation value sequence:

[0102] The first sub-step, in response to determining that the image change information of the previous frame target coded image after denoising is, that the content of the previous frame target coded image after denoising has changed compared to the previous frame target coded image after denoising, and that the first compensation deviation value is greater than or equal to the fifth preset threshold, determines the first change value as the sum of the preset change value and the first preset value. Wherein, the previous frame target coded image after denoising is the target coded image of the previous frame target coded image after denoising. Wherein, the fifth preset threshold can be a single value. The fifth preset threshold can be any value in [30, 66].

[0103] The second sub-step, in response to determining that the image change information of the previous frame target coded image after noise reduction is information indicating a change in content between the previous frame target coded image after noise reduction and the previous frame target coded image after noise reduction, and that the first compensation deviation value is less than the fifth preset threshold and the first compensation deviation value is greater than or equal to the ratio of the fifth preset threshold to the third preset value, the sum of the ratio of the fifth preset threshold to the third preset value and the first compensation deviation value is determined as the first change deviation value, and the sum of the preset change value and the first preset value is determined as the first change value. The third preset value can be a single numerical value. The third preset value can be the value "2".

[0104] The third sub-step, in response to determining that the image change information of the previous frame target coded image after noise reduction is information indicating a change in content between the previous frame target coded image after noise reduction and the previous frame target coded image after noise reduction, and that the first compensation deviation value is less than the fifth preset threshold, less than the ratio of the fifth preset threshold to the third preset value, and greater than or equal to the ratio of the fifth preset threshold to the fourth preset value, determines the product of the first compensation deviation value and the third preset value as the first change deviation value. The fourth preset value can be a single numerical value. For example, the fourth preset value can be the value "4".

[0105] The fourth sub-step is to determine, in response to the determination that the image change information of the previous frame target coded image after noise reduction is information that the content of the previous frame target coded image after noise reduction has not changed compared with the previous frame target coded image after noise reduction, or the image change information of the previous frame target coded image after noise reduction is information that a small amount of content has changed compared with the previous frame target coded image after noise reduction, and the first compensation deviation value is greater than or equal to the fifth preset threshold, the sum of the preset change value and the first preset value is determined as the first provisional change value.

[0106] The fifth sub-step, in response to the completion of the determination step, determines the aforementioned first provisional change value as the first change value.

[0107] The sixth sub-step, in response to the determination step not being completed, sets the first provisional change value as a preset change value, and executes the determination step again.

[0108] In some optional implementations of certain embodiments, the first change value and the second change value are subjected to a second change processing to obtain image change information of the target encoded image. This image change information may be information about changes in content between the target encoded image and the previous frame target encoded image after noise reduction. This may include the following steps:

[0109] The first step is to determine the third change value by summing the first change value and the second change value.

[0110] The second step involves determining that, in response to the determination that the third change value is greater than the sixth preset threshold, the image change information of the target coded image is defined as information indicating that the content of the target coded image has changed compared to the previous frame target coded image after noise reduction. The sixth preset threshold can be a numerical value. Specifically, the sixth preset threshold can be any value in the range [12, 32].

[0111] Third step: In response to determining that the third change value is less than or equal to the sixth preset threshold, and that the image change information of the previous frame target coded image after noise reduction is information that the content of the previous frame target coded image after noise reduction has changed compared with the previous frame target coded image after noise reduction, and that the third change value is greater than the ratio of the sixth preset threshold to the third preset value, the image change information of the target coded image is determined to be information that the content of the target coded image has changed compared with the previous frame target coded image after noise reduction.

[0112] Fourth step, in response to determining that the third change value is less than or equal to the sixth preset threshold, and that the image change information of the previous frame target coded image after noise reduction is information that a small amount of content has changed compared to the previous frame target coded image after noise reduction, and that the third change value is less than or equal to the ratio of the sixth preset threshold to the third preset value, the image change information of the target coded image is determined to be information that a small amount of content has changed compared to the previous frame target coded image after noise reduction.

[0113] Fifth step: In response to determining that the third change value is less than or equal to the second preset value, the image change information of the target coded image is determined to be information that the content of the target coded image does not change compared with the previous frame target coded image after noise reduction.

[0114] In some optional implementations of certain embodiments, the above-described compensation processing of the first overall deviation value and each first pixel deviation value in the first pixel deviation value sequence to obtain the first compensated deviation value sequence may include the following steps:

[0115] The first step involves determining, in response to the determination that the first overall deviation value is less than or equal to a preset minimum value, the seventh preset threshold is set as the adjustment multiple. The seventh preset threshold can be a single numerical value. For example, the seventh preset threshold can be any value in the range [1, 4]. The preset minimum value can also be a single numerical value. For example, the preset minimum value can be the value "0".

[0116] The second step involves determining that the first overall deviation value is greater than a preset minimum value and less than a preset maximum value, and then setting the determined multiplier as the adjustment multiplier. The preset maximum value can be a single numerical value. For example, the preset maximum value could be the value "64". The determined multiplier is obtained through the following steps:

[0117] The first sub-step is to determine the difference between the aforementioned seventh preset threshold and the first preset value as the first multiple value.

[0118] The second sub-step involves determining the difference between the aforementioned preset maximum value and the aforementioned first overall deviation value as a second multiple value.

[0119] The third sub-step involves determining the difference between the aforementioned preset maximum value and the aforementioned preset minimum value as the third multiple value.

[0120] The fourth sub-step is to determine the fourth multiple value by combining the third multiple value with the first preset value.

[0121] The fifth sub-step involves multiplying the first and second multiple values ​​to determine the fifth multiple value.

[0122] The sixth sub-step is to determine the ratio of the fifth multiple value to the fourth multiple value as the determined multiple value.

[0123] The third step is to multiply the third comprehensive deviation value by the above adjustment multiple value and determine it as the provisional compensation deviation value.

[0124] Fourth step: In response to determining that the first overall deviation value is greater than the eighth preset threshold, the sum of the fifth preset threshold and the provisional compensation deviation value is determined as the first compensation deviation value. The eighth preset threshold can be a numerical value. For example, the eighth preset threshold can be any value in [200, 255].

[0125] Fifth step: In response to determining that the first overall deviation value is less than or equal to the eighth preset threshold and the first overall deviation value is less than the ninth preset threshold, or the provisional compensation deviation value is less than the ratio of the fifth preset threshold to the fourth preset value, the sum of the ratio of the fifth preset threshold to the third preset value and the provisional compensation deviation value is determined as the first compensation deviation value. The ninth preset threshold can be a single numerical value. For example, the ninth preset threshold can be any value in the range [0, 20].

[0126] The above-mentioned content, as an inventive point of this disclosure, solves the third technical problem mentioned in the background art: "The brightness of dark areas in the target image after noise reduction is low, and the brightness of dark areas cannot be compensated, resulting in low image clarity." Factors leading to low image clarity often include: the brightness of dark areas in the target image after noise reduction is low, and the brightness of dark areas cannot be compensated, resulting in low image clarity. Solving these factors can achieve a high accuracy rate for the danger value. To achieve this effect, firstly, in response to determining that the first overall deviation value is less than or equal to a preset minimum value, a seventh preset threshold is determined as an adjustment multiple. The seventh preset threshold can be a single value. For example, the seventh preset threshold can be any value in [1, 4]. The preset minimum value can be a single value. For example, the preset minimum value can be the value "0". Secondly, in response to determining that the first overall deviation value is greater than the preset minimum value and the first overall deviation value is less than a preset maximum value, the determined multiple is determined as an adjustment multiple. The preset maximum value can be a single value. For example, the preset maximum value can be the value "64". The aforementioned multiplier value is obtained through the following steps: First, the difference between the seventh preset threshold and the first preset value is determined as the first multiplier value. Second, the difference between the aforementioned preset maximum value and the aforementioned first overall deviation value is determined as the second multiplier value. Then, the sum of the difference between the aforementioned preset maximum value and the aforementioned preset minimum value and the first preset value is determined as the third multiplier value. Finally, the ratio of the product of the aforementioned first multiplier value and the second multiplier value to the aforementioned third multiplier value is determined as the final multiplier value. Third, the product of the third comprehensive deviation value and the aforementioned adjustment multiplier value is determined as the provisional compensation deviation value. The provisional compensation deviation value can be the deviation value after compensating for dark areas. Since the overall brightness of dark areas in an image is relatively low, changes in the overall dark areas of the image are not easily noticeable, which can avoid motion blur when pixels are superimposed, thereby improving image clarity. Fourth, in response to determining that the aforementioned first overall deviation value is greater than the eighth preset threshold, the sum of the fifth preset threshold and the aforementioned provisional compensation deviation value is determined as the first compensation deviation value. The aforementioned eighth preset threshold can be a single numerical value. For example, the eighth preset threshold can be any value in [200, 255]. Fifth, in response to determining that the first overall deviation value is less than or equal to the eighth preset threshold and the first overall deviation value is less than the ninth preset threshold, or the provisional compensation deviation value is less than the ratio of the fifth preset threshold to the fourth preset value, the sum of the ratio of the fifth preset threshold to the third preset value and the provisional compensation deviation value is determined as the first compensation deviation value. The ninth preset threshold can be a single value. For example, the ninth preset threshold can be any value in [0, 20]. The first compensation deviation value can be the deviation value after compensating for overexposure or underexposure.Overexposure and underexposure can cause image ghosting and affect whether the target encoded image of the current frame has changed compared to the previous frame's target encoded image after noise reduction. Therefore, by determining whether an image is overexposed or underexposed, corresponding deep compensation can be applied. This avoids affecting whether the target encoded image of the current frame has changed compared to the previous frame's target encoded image after noise reduction and improves image sharpness.

[0127] Step 1024: Weight each third pixel deviation value in the third pixel deviation value sequence with the corresponding value of the target coded pixel in the target coded image to generate the denoised target coded pixel, and obtain the denoised target coded image.

[0128] In some embodiments, each third pixel deviation value in the third pixel deviation value sequence is weighted with the value corresponding to the target coded pixel in the target coded image to generate a denoised target coded pixel, thereby obtaining a denoised target coded image.

[0129] In practice, the process of weighting each third pixel deviation value in the third pixel deviation value sequence with the corresponding value of the target coded pixel in the target coded image to generate a denoised target coded pixel and thus a denoised target coded image can include the following steps:

[0130] The first step is to visualize the above-mentioned third pixel deviation value sequence and the preset weighted threshold to obtain the preset weighted threshold corresponding to each third pixel deviation value in the above-mentioned third pixel deviation value sequence. The preset weighted threshold value is in the range [0,1].

[0131] The second step involves performing the following steps for each preset weighted threshold corresponding to the third pixel deviation value in the above sequence of third pixel deviation values:

[0132] The first sub-step involves determining the first weighting value as the product of the preset weighting threshold and the target encoded pixel corresponding to the third pixel deviation value.

[0133] The second sub-step involves determining the second weighting value as the product of the difference between 1 and the aforementioned preset weighting threshold and the product of the denoised target encoded pixel of the previous frame corresponding to the aforementioned third pixel deviation value.

[0134] The third sub-step involves determining the sum of the first weighted value and the second weighted value as the target encoded pixel after noise reduction.

[0135] The third step is to determine the multiple denoised target encoded pixels as the denoised target encoded image.

[0136] Optionally, after performing the above noise reduction processing on the target encoded image in the above target encoded image sequence, the executing entity may further include the following steps:

[0137] The first step involves determining that the target coded image is the first frame target coded image, and then defining the first frame target coded image as the denoised target coded image. The image change information in the denoised target coded image refers to information indicating that the content of the target coded image has not changed compared to the previous denoised target coded image.

[0138] The second step is to determine, in response to the determination that the above target coded image is the last frame target coded image and that the above target coded image is the denoised target coded image, to determine each denoised target coded image as the denoised target coded video.

[0139] The above-described embodiments of this disclosure have the following beneficial effects: the image denoising processing method of some embodiments of this disclosure can improve the quality of the denoised target encoded image, thereby improving the quality of the denoised target encoded video. Specifically, the reason for the low quality of the denoised target image is that: when using a preset denoising convolutional neural network to denoise the target image, it is impossible to denoise based on the previous frame of the target image, thus resulting in low quality of the denoised target image. Based on this, some embodiments of the image denoising processing method disclosed herein obtain a target coded image sequence corresponding to a target coded video, wherein the format of the target coded video is a chroma and luminance format, and the chroma and luminance format includes a first chroma format, a second chroma format, a first luminance format, and a second luminance format; for the target coded image in the target coded image sequence, in response to determining that the target coded image is not the first frame target coded image and that the target coded image is an undenoised target coded image, the following denoising processing is performed: based on the target coded image and the denoised previous frame target coded image corresponding to the target coded image, a first luminance deviation value sequence and a second luminance deviation value sequence are determined; wherein the first luminance deviation value in the first luminance deviation value sequence may be the deviation value between the target coded pixel of the first luminance format in the target coded image and the denoised target coded pixel of the first luminance format in the denoised previous frame target coded image. The second brightness deviation value in the aforementioned second brightness deviation value sequence can be the deviation value between the target coded pixel in the second brightness format in the aforementioned target coded image and the denoised target coded pixel in the second brightness format of the aforementioned denoised previous frame target coded image. The first brightness deviation value and the second brightness deviation value are determined using the aforementioned target coded image and the aforementioned denoised previous frame target coded image, resulting in more accurate first and second brightness deviation values. Based on the aforementioned first brightness deviation value sequence and the aforementioned second brightness deviation value sequence, a first pixel deviation value sequence and a second pixel deviation value sequence are generated. The first pixel deviation value in the aforementioned first pixel deviation value sequence can be the deviation value between the target coded pixel in the first chroma format in the aforementioned target coded image and the denoised target coded pixel in the first chroma format of the aforementioned denoised previous frame target coded image, and the deviation value of the first brightness deviation value in the aforementioned first brightness deviation value sequence. The second pixel deviation value in the above second pixel deviation value sequence may be the deviation value of the target coded pixel in the second chroma format in the above target coded image, the deviation value of the target coded pixel in the second chroma format of the above denoised previous frame target coded image, and the deviation value of the second brightness deviation value in the above second brightness deviation value sequence.Based on the first pixel deviation value sequence and the second pixel deviation value sequence, a third pixel deviation value sequence is generated. The third pixel deviation value can be obtained by stitching together the first and second pixel deviation values ​​after exposure compensation. Each third pixel deviation value in the third pixel deviation value sequence is weighted with the corresponding value of the target coded pixel in the target coded image to generate a denoised target coded pixel, resulting in a denoised target coded image. By visualizing the data, a preset weighted value corresponding to the third pixel deviation value is obtained, and weighted with the corresponding target coded pixel to obtain the denoised target coded pixel. This improves the quality of the denoised target coded image, thereby improving the quality of the denoised target coded video.

[0140] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of an image noise reduction processing apparatus, which are similar to... Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0141] like Figure 2 As shown, the image noise reduction processing apparatus 200 in some embodiments includes an acquisition unit 201 and an execution unit 202. The acquisition unit 201 is configured to acquire a target encoded image sequence corresponding to a target encoded video, wherein the target encoded video is in chroma and luminance format, including a first chroma format, a second chroma format, a first luminance format, and a second luminance format. The extraction processing unit 202 is configured to, for a target encoded image in the target encoded image sequence, in response to determining that the target encoded image is not the first frame target encoded image and that the target encoded image is an undenoised target encoded image, perform the following denoising processing: determining a first luminance deviation value sequence and a second luminance deviation value sequence based on the target encoded image and the corresponding denoised previous frame target encoded image; generating a first pixel deviation value sequence and a second pixel deviation value sequence based on the first luminance deviation value sequence and the second luminance deviation value sequence; generating a third pixel deviation value sequence based on the first pixel deviation value sequence and the second pixel deviation value sequence; and weighting each third pixel deviation value in the third pixel deviation value sequence with the value corresponding to the target encoded pixel in the target encoded image to generate a denoised target encoded pixel, thereby obtaining a denoised target encoded image.

[0142] It is understandable that the units described in the image noise reduction processing apparatus 200 are related to the reference... Figure 1The steps in the described method correspond to each other. Therefore, the operations, features, and beneficial effects described above for the method also apply to the device 200 and the units contained therein, and will not be repeated here.

[0143] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0144] like Figure 3 As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0145] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0146] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0147] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0148] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0149] The aforementioned computer-readable medium may be included in the aforementioned device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire a target encoded image sequence corresponding to a target encoded video, wherein the target encoded video is in chroma and luma format, the chroma and luma format including a first chroma format, a second chroma format, a first luma format, and a second luma format; for a target encoded image in the target encoded image sequence, in response to determining that the target encoded image is not the first frame target encoded image and that the target encoded image is an undenoised target encoded image, perform the following denoising processing: according to the target encoded... The target encoded image corresponds to the previous frame target encoded image after denoising. A first brightness deviation value sequence and a second brightness deviation value sequence are determined. Based on the first brightness deviation value sequence and the second brightness deviation value sequence, a first pixel deviation value sequence and a second pixel deviation value sequence are generated. Based on the first pixel deviation value sequence and the second pixel deviation value sequence, a third pixel deviation value sequence is generated. Each third pixel deviation value in the third pixel deviation value sequence is weighted with the value corresponding to the target encoded pixel in the target encoded image to generate the denoised target encoded pixel, thus obtaining the denoised target encoded image.

[0150] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0151] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0152] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including an acquisition unit and an execution unit. The names of these units do not necessarily limit the unit itself; for example, an acquisition unit may also be described as "a unit that acquires a target coded image sequence corresponding to a target coded video."

[0153] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0154] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. An image noise reduction processing method, comprising: Obtain the target encoded image sequence corresponding to the target encoded video, wherein the format of the target encoded video is chroma and luma format, and the chroma and luma format includes a first chroma format, a second chroma format, a first luma format, and a second luma format; For a target coded image in the target coded image sequence, in response to determining that the target coded image is not the first frame target coded image and that the target coded image is an undenoised target coded image, the following denoising processing is performed: Based on the target coded image and the previous frame target coded image after noise reduction, a first brightness deviation value sequence and a second brightness deviation value sequence are determined. Based on the first brightness deviation value sequence and the second brightness deviation value sequence, a first pixel deviation value sequence and a second pixel deviation value sequence are generated; A third pixel deviation value sequence is generated based on the first pixel deviation value sequence and the second pixel deviation value sequence; Each third pixel deviation value in the third pixel deviation value sequence is weighted with the value corresponding to the target coded pixel in the target coded image to generate a denoised target coded pixel, thus obtaining a denoised target coded image.

2. The method according to claim 1, wherein, The step of determining the first brightness deviation value sequence and the second brightness deviation value sequence based on the target coded image and the corresponding denoised previous frame target coded image includes: The difference between the average pixel value in the first pixel block and the average pixel value in the second pixel block is determined as the first deviation value. The previous frame target encoded image after noise reduction includes the first pixel block and the target encoded image includes the second pixel block. The first pixel block is composed of pixels of the first chroma format in the chroma and luminance formats, and the second pixel block is composed of pixels of the first chroma format in the chroma and luminance formats. The difference between the average pixel value in the third pixel block and the average pixel value in the fourth pixel block is determined as the second deviation value. The previous frame target encoded image after noise reduction includes a third pixel block and a fourth pixel block. The third pixel block is composed of pixels of the second chroma format in the chroma and luminance formats. The fourth pixel block is composed of pixels of the second chroma format in the chroma and luminance formats. The sum of the first absolute value and the second absolute value is determined as the color deviation value, wherein the first absolute value is the absolute value of the first deviation value, and the second absolute value is the absolute value of the second deviation value; The fifth pixel block is divided into an average number of first sub-pixel blocks to obtain a preset number of first sub-pixel blocks, which are used as the first sub-pixel block sequence. The previous frame target encoded image after noise reduction includes the fifth pixel block, which is composed of multiple pixels of the first brightness format in the previous frame target encoded image after noise reduction. The sixth pixel block is divided into an average number of second sub-pixel blocks to obtain a predetermined number of second sub-pixel blocks, which are used as a second sub-pixel block sequence. The target encoded image includes the sixth pixel block, which is composed of multiple pixels of the first brightness format in the target encoded image. The seventh pixel block is divided into an average number of third sub-pixel blocks to obtain a preset number of third sub-pixel blocks as a sequence of third sub-pixel blocks. The previous frame target coded image after noise reduction includes the seventh pixel block, which is composed of multiple pixels of the second brightness format in the previous frame target coded image after noise reduction. The eighth pixel block is divided into an average number of fourth sub-pixel blocks to obtain a predetermined number of fourth sub-pixel blocks, which are used as a sequence of fourth sub-pixel blocks. The target encoded image includes an eighth pixel block, which is composed of multiple pixels of the second brightness format in the target encoded image. Each first sub-pixel block in the first sub-pixel block sequence and the corresponding second sub-pixel block in the second sub-pixel block sequence are subjected to a first preset average value to generate a first brightness deviation value, thereby obtaining a first brightness deviation value sequence, wherein there is a one-to-one correspondence between the first sub-pixel block in the first sub-pixel block sequence and the second sub-pixel block in the second sub-pixel block sequence; Each third sub-pixel block in the third sub-pixel block sequence and its corresponding fourth sub-pixel block in the fourth sub-pixel block sequence are subjected to a second preset average value processing to generate a second brightness deviation value, thereby obtaining a second brightness deviation value sequence, wherein there is a one-to-one correspondence between the third sub-pixel blocks in the third sub-pixel block sequence and the fourth sub-pixel blocks in the fourth sub-pixel block sequence.

3. The method according to claim 1, wherein, The step of generating a third pixel deviation value sequence based on the first pixel deviation value sequence and the second pixel deviation value sequence includes: Compensation processing is performed on the first overall deviation value and each first pixel deviation value in the first pixel deviation value sequence to obtain the first compensated deviation value sequence. The average value of the first brightness deviation value sequence is determined as the first overall deviation value. Compensation processing is performed on the second overall deviation value and each second pixel deviation value in the second pixel deviation value sequence to obtain the second compensated deviation value sequence. The average value of the second brightness deviation value sequence is determined as the second overall deviation value. The first compensation deviation value sequence is subjected to a first change process to obtain a first change value and a first change deviation value; The second compensation deviation value sequence is subjected to a first change process to obtain a second change value and a second change deviation value; A second change processing is performed on the first change value and the second change value to obtain the image change information of the target encoded image; The obtained first and second variation deviation value sequences are concatenated to obtain the third pixel deviation value sequence.

4. The method according to claim 3, wherein, The first transformation process performed on the first compensation deviation value sequence to obtain a first transformation value and a first transformation deviation value includes: The second preset value is set as the preset change value; For each first compensation deviation value in the first compensation deviation value sequence, the following determination steps are performed: In response to determining that the image change information of the previous frame target coded image after noise reduction is information that the content of the previous frame target coded image after noise reduction has changed compared with the previous frame target coded image after noise reduction, and the first compensation deviation value is greater than or equal to the fifth preset threshold, the sum of the preset change value and the first preset value is determined as the first change value, wherein the previous frame target coded image after noise reduction is the target coded image of the previous frame after noise reduction; In response to determining that the image change information of the previous frame target coded image after noise reduction is information that the content of the previous frame target coded image after noise reduction has changed compared with the previous frame target coded image after noise reduction, and that the first compensation deviation value is less than the fifth preset threshold, and that the first compensation deviation value is greater than or equal to the ratio of the fifth preset threshold to the third preset value, the sum of the ratio of the fifth preset threshold to the third preset value and the first compensation deviation value is determined as the first change deviation value, and the sum of the preset change value and the first preset value is determined as the first change value; In response to determining that the image change information of the previous frame target coded image after noise reduction is information that the content of the previous frame target coded image after noise reduction has changed compared with the previous frame target coded image after noise reduction, and the first compensation deviation value is less than the fifth preset threshold, and the first compensation deviation value is less than the ratio of the fifth preset threshold to the third preset value, and the first compensation deviation value is greater than or equal to the ratio of the fifth preset threshold to the fourth preset value, the product of the first compensation deviation value and the third preset value is determined as the first change deviation value; In response to determining that the image change information of the previous frame target coded image after noise reduction is that the content of the previous frame target coded image after noise reduction has not changed compared with the previous frame target coded image after noise reduction, or that the image change information of the previous frame target coded image after noise reduction is that a small amount of content has changed compared with the previous frame target coded image after noise reduction, and the first compensation deviation value is greater than or equal to the fifth preset threshold, the sum of the preset change value and the first preset value is determined as the first provisional change value; In response to the determination step being completed, the first provisional change value is determined as the first change value; In response to the determination that the determination step has not been completed, the first provisional change value is determined as a preset change value, so that the determination step is executed again.

5. The method according to claim 4, wherein, The second change processing of the first change value and the second change value to obtain the image change information of the target encoded image includes: The sum of the first change value and the second change value is determined as the third change value; In response to determining that the third change value is greater than the sixth preset threshold, the image change information of the target coded image is determined as information that the content of the target coded image has changed compared with the previous frame target coded image after noise reduction; In response to determining that the third change value is less than or equal to the sixth preset threshold, and that the image change information of the previous frame target coded image after noise reduction is information that the content of the previous frame target coded image after noise reduction has changed compared with the previous frame target coded image after noise reduction, and that the third change value is greater than the ratio of the sixth preset threshold to the third preset value, the image change information of the target coded image is determined to be information that the content of the target coded image has changed compared with the previous frame target coded image after noise reduction; In response to determining that the third change value is less than or equal to the sixth preset threshold, and that the image change information of the previous frame target coded image after noise reduction is information that a small amount of content has changed compared to the previous frame target coded image after noise reduction, and that the third change value is less than or equal to the ratio of the sixth preset threshold to the third preset value, the image change information of the target coded image is determined to be information that a small amount of content has changed compared to the previous frame target coded image after noise reduction; In response to determining that the third change value is less than or equal to the second preset value, the image change information of the target coded image is determined to be information that the content of the target coded image does not change compared with the previous frame target coded image after noise reduction.

6. The method according to claim 1, wherein, After performing the following noise reduction processing on the target encoded images in the target encoded image sequence, the method further includes: In response to determining that the target encoded image is a first frame target encoded image, the first frame target encoded image is determined as a denoised target encoded image, wherein the image change information of the denoised target encoded image is information that the content of the target encoded image has not changed compared with the previous frame target encoded image after denoising; In response to determining that the target coded image is the last frame of the target coded image and that the target coded image is a denoised target coded image, each denoised target coded image is determined as a denoised target coded video.

7. An image noise reduction processing apparatus, comprising: The acquisition unit is configured to acquire a target encoded image sequence corresponding to a target encoded video, wherein the target encoded video is in chroma and luma format, and the chroma and luma format includes a first chroma format, a second chroma format, a first luma format, and a second luma format; An execution unit is configured to, for a target encoded image in the target encoded image sequence, in response to determining that the target encoded image is not the first frame target encoded image and that the target encoded image is an undenoised target encoded image, perform the following denoising processing: determining a first brightness deviation value sequence and a second brightness deviation value sequence based on the target encoded image and the corresponding denoised previous frame target encoded image; generating a first pixel deviation value sequence and a second pixel deviation value sequence based on the first brightness deviation value sequence and the second brightness deviation value sequence; generating a third pixel deviation value sequence based on the first pixel deviation value sequence and the second pixel deviation value sequence; and weighting each third pixel deviation value in the third pixel deviation value sequence with the value corresponding to the target encoded pixel in the target encoded image to generate a denoised target encoded pixel, thereby obtaining a denoised target encoded image.

8. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-6.

9. A computer-readable medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

Images