Image processing method and device, electronic equipment and computer readable storage medium

Abstract

The disclosure provides an image processing method and device, electronic equipment and computer readable storage medium, relating to the technical field of image processing. The method comprises: acquiring an image set, the image set comprising a first image and at least one frame of second image, the first image and the second image being images of different brightness of the same shooting scene; dividing the first image into a plurality of initial image blocks, and determining a set of similar image blocks corresponding to each initial image block from the image set; performing frequency domain conversion on each similar image block in the set of similar image blocks to determine the frequency component of each similar image block; fusing the frequency components of each similar image block, and performing spatial domain conversion on the result after frequency component fusion to obtain a noise-reduced image block corresponding to the initial image block; and generating a noise-reduced image according to the noise-reduced image block corresponding to each initial image block. The disclosure can improve the noise reduction processing effect of the image.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of image processing, in particular to an image processing method, an image processing apparatus, an electronic device and a computer readable storage medium. BACKGROUND

[0002] With the rapid development of imaging technology, the function of imaging equipment is becoming more and more powerful, and the user's requirement for image quality is gradually increasing. However, due to various reasons, there are often some noise points in the image, which will affect the image quality.

[0003] At present, the image denoising method of the related technology has the defect that the denoising effect is not ideal, which affects the user's perception to some extent. SUMMARY

[0004] The purpose of the present disclosure is to provide an image processing method, an image processing apparatus, an electronic device and a computer readable storage medium, thereby at least to some extent improving the denoising effect of the image.

[0005] According to a first aspect of the present disclosure, an image processing method is provided, comprising: acquiring an image set, the image set comprising a first image and at least one frame of second image, the first image and the second image being images of different brightness of the same shooting scene; dividing the first image into a plurality of initial image blocks, and determining a set of similar image blocks corresponding to each initial image block from the image set; performing frequency domain conversion on each similar image block in the set of similar image blocks to determine the frequency component of each similar image block; fusing the frequency components of each similar image block, and performing spatial domain conversion on the result after fusing the frequency components to obtain a denoising image block corresponding to the initial image block; and generating a denoised image according to the denoising image block corresponding to each initial image block.

[0006] According to a second aspect of the present disclosure, an image processing apparatus is provided, comprising: an acquisition module configured to acquire an image set, the image set comprising a first image and at least one frame of second image, the first image and the second image being images of different brightness of the same shooting scene; a first processing module configured to divide the first image into a plurality of initial image blocks, and determine a set of similar image blocks corresponding to each initial image block from the image set; an image conversion module configured to perform frequency domain conversion on each similar image block in the set of similar image blocks to determine the frequency component of each similar image block; a second processing module configured to fuse the frequency components of each similar image block, and perform spatial domain conversion on the result after fusing the frequency components to obtain a denoising image block corresponding to the initial image block; and a third processing module configured to generate a denoised image according to the denoising image block corresponding to each initial image.

[0007] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the method described above.

[0008] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, which stores a computer program, which when executed by a processor, implements the method described above.

[0009] The image processing method provided by the embodiments of the present disclosure divides the first image into a plurality of initial image blocks, determines a set of similar image blocks corresponding to each initial image block from the image set, converts the similar image blocks in the set of similar image blocks from a spatial domain to a frequency domain to obtain frequency components of the similar image blocks, fuses the frequency components of the similar image blocks, and converts the result after the fusion of the frequency components from the frequency domain to the spatial domain to obtain a denoised image block corresponding to the initial image block. The denoised image is generated according to the denoised image blocks corresponding to all the initial image blocks. On the one hand, the image set includes images of different brightness of the same shooting scene, so that the set of similar image blocks corresponding to the initial image block increases the image blocks of different brightness, increases the reference information of the denoising process, and enriches the image brightness information at each frequency component fusion stage, which helps to realize high dynamic range imaging. On the other hand, the frequency components of the similar image blocks are fused, so that the denoised image block increases the texture information corresponding to the high frequency components from the similar image blocks and the image flat area information corresponding to the low frequency components, and improves the performance of the denoised image block in high frequency and low frequency.

[0010] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0011] The drawings incorporated into the specification and forming a part thereof, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure. It is clear that the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained from these drawings without creative labor for those skilled in the art. In the drawings:

[0012] Figure 1 A schematic diagram showing an exemplary application environment related to an image processing method and device to which the embodiments of the present disclosure can be applied is shown;

[0013] Figure 2 A schematic diagram showing an exemplary application environment related to another image processing method and device in an exemplary embodiment of the present disclosure is shown;

[0014] Figure 3A flow chart of an image processing method in an example embodiment of the present disclosure is schematically shown;

[0015] Figure 4 A schematic diagram of a way of obtaining a similar image block in an example embodiment of the present disclosure is schematically shown;

[0016] Figure 5 A schematic diagram of a way of calculating the similarity between the neighborhood blocks in a non-local mean filtering manner in an example embodiment of the present disclosure is schematically shown;

[0017] Figure 6 A schematic diagram of a way of converting the similar image block to the frequency domain in an example embodiment of the present disclosure is schematically shown;

[0018] Figure 7 A schematic diagram of an example of an image processing method in the present disclosure is schematically shown;

[0019] Figure 8 A flow chart of an implementation of a way of determining a set of similar image blocks corresponding to an initial image block in an example embodiment of the present disclosure is schematically shown;

[0020] Figure 9 A flow chart of a way of obtaining a denoised image block corresponding to an initial image block in an example embodiment of the present disclosure is schematically shown;

[0021] Figure 10 A schematic diagram of a way of obtaining a corresponding relationship between the noise intensity ratio and the noise intensity in an example embodiment of the present disclosure is schematically shown;

[0022] Figure 11 A flow chart of an image processing method in an application scenario related to an example embodiment of the present disclosure is schematically shown;

[0023] Figure 12 A schematic diagram of the composition of an image processing device in an example embodiment of the present disclosure is schematically shown;

[0024] Figure 13 A schematic diagram of an electronic device to which an embodiment of the present disclosure can be applied is shown. DETAILED DESCRIPTION

[0025] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art. Features described in the description, structures, or characteristics may be combined in any suitable manner in one or more implementations.

[0026] In addition, the accompanying drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0027] Figure 1 A schematic diagram shows an exemplary application environment involved in an image processing method and apparatus according to an embodiment of the present disclosure.

[0028] like Figure 1 As shown, the terminal device can be an intelligent device with image processing capabilities, such as a smart phone, computer, tablet computer, smart watch, vehicle-mounted device, wearable device, monitoring device and other intelligent devices. The terminal device can also be called a mobile terminal, terminal, mobile device, etc. The present disclosure does not limit the type of terminal device.

[0029] The images in the image set are images of the same captured scene at different brightness levels, with the first image being the image to be denoised. The images in the image set may be captured by a terminal device or acquired from a network or other device. The exposure durations of the images in the image set may be the same or different, with no particular limitation. In the disclosed embodiment, the image processing device of the terminal device may perform denoising on the first image based on the image processing method of the disclosed embodiment to obtain a denoised image.

[0030] It should be noted that the image processing method provided in the embodiment of the present disclosure can be executed by a terminal device. The image processing method provided in the embodiment of the present disclosure can also be executed by a server, and accordingly, the image processing device can also be set in the server. Figure 2 As shown, the terminal device can send an image set to the server, so that the server performs noise reduction processing on the first image based on the image processing method of the embodiment of the present disclosure, and sends the noise-reduced image to the terminal device. The server can be a background system that provides image processing related services in the embodiment of the present disclosure, and can include a single electronic device with computing capabilities such as a portable computer, desktop computer, or smartphone, or a cluster of multiple electronic devices. In the embodiment of the present disclosure, the image processing method is executed by the terminal device as an example for description.

[0031] In the related art, in the process of collecting, transmitting or storing images, the images can generate noise due to various reasons, affecting the image quality, and thus the images need to be denoised. For example, images of different exposure durations can be fused and then denoised, or images of different exposure durations can be denoised respectively and then fused. However, the correlation between the images is not considered, the image texture information and the smoothness of the denoised image cannot be considered at the same time, the denoising effect is not good, and the image visual effect is affected.

[0032] Based on one or more of the above problems, the exemplary embodiments of the present disclosure provide an image processing method. Referring to Figure 3 The image processing method can include the following steps S310 to S350 as shown in the figure:

[0033] In step S310, an image set is acquired, the image set including a first image and at least one second image, the first image and the second image being images of different brightness of the same shooting scene.

[0034] In the exemplary embodiments of the present disclosure, the images in the image set can be obtained by a terminal device shooting the same shooting scene, can all come from other devices or networks, or part of them are obtained by the terminal device shooting and the other part comes from other devices or networks, etc.; the images in the image set can be multiple frames of images of different brightness obtained at the same exposure time, or can be multiple frames of images of different brightness obtained in different exposure times for the same shooting scene, and no special limitation is made to this.

[0035] In some possible embodiments, the DCG (dual conversion gain) technology can be used to output two frames of images of different brightness at the same exposure time, and since the exposure times of the two frames of images are synchronized, the image contents are basically aligned. For example, based on the double exposure or multiple exposure technology, multiple frames of images of different exposure times are obtained in a short time, the multiple frames of images are of different brightness, etc., and the present disclosure does not make special limitation to the image collection method.

[0036] The first image is a to-be-denoised image in the image set. For example, the image set includes image A and image B, and image A can be denoised based on the image set, or image B can be denoised according to the image set, or a fused image of image A and image B can be denoised according to the image set, that is, in the present disclosure, the first image can be any image or a fused image corresponding to any image pair in the image set. Since the first image and the second image are images of different brightness of the same shooting scene, the noise intensities of the two images are different, and the image contents can be represented to different degrees.

[0037] In step S320, the first image is divided into a plurality of initial image blocks, and a set of similar image blocks corresponding to each initial image block is determined from the image set.

[0038] In the example embodiment of the present disclosure, the first image can be divided into a plurality of initial image blocks according to actual noise reduction requirements, for example, divided into 8x8, 16x16 pixel blocks, and no special requirements are made thereon, so as to respectively reduce noise of each initial image block to obtain a noise-reduced image block. The first image is divided and noise-reduced, the local characteristics of the first image can be fully utilized, and the calculation complexity is reduced.

[0039] There are many ways to measure the similarity index between the initial image block and the similar image block, for example, calculating the mean square error (MSE) and the like.

[0040] In some possible embodiments, the set of similar image blocks can include similar image blocks from the first image and similar image blocks from the second image.

[0041] In some possible embodiments, the set of similar image blocks can only include similar image blocks from the second image, and no special limitation is made thereon.

[0042] In some possible embodiments, the set of similar image blocks can further include the initial image block itself.

[0043] Figure 4 A schematic diagram of a way of obtaining similar image blocks by using a neighborhood window according to an example embodiment of the present disclosure is shown. The following takes determining similar image blocks in the first image as an example for description.

[0044] Referring to Figure 4 The large window is a search window centered on the center pixel (target pixel x) of the initial image block, and the small windows are neighborhood windows centered on the target pixel x and the pixel y, respectively. The neighborhood window centered on the pixel y slides in the search window, and the similarity between the two neighborhood windows is calculated to assign a value to the pixel y. As shown in Figure 5 A schematic diagram of calculating the similarity between the neighborhood block Y centered on the pixel y and the neighborhood block X centered on the target pixel x in a non-local mean filtering manner is shown. Based on the pixel pairs formed by the pixels in the neighborhood block Y and the pixels in the neighborhood block X, the similarity between the two neighborhood blocks is calculated in a mean square error manner.

[0045] It is worth noting that, Figure 4 And Figure 5 are only examples, and the embodiment of the present disclosure can also select a method of obtaining similar image blocks according to actual image processing requirements, such as a Gaussian filter space method, etc.

[0046] The similar image block set is formed according to the obtained similar image blocks, and the similar image block set of each initial image block is obtained, so that the redundant information in each image with different brightness is fully utilized by combining the sparse and low-rank properties between the similar image blocks, and the details of the image are maintained to the greatest extent while denoising.

[0047] In step S330, the frequency components of each similar image block in the similar image block set are converted in the frequency domain.

[0048] In the exemplary embodiments of the present disclosure, the frequency domain conversion is to convert the similar image block from the spatial domain to the frequency domain, and the high-frequency component part in the frequency domain can reflect the image details, edges, noise information and the like, and the low-frequency component part can reflect the information of the flat area of the image.

[0049] After the image in the spatial domain is converted to the frequency domain, the frequency components of the similar image block can be obtained. For example, Figure 6 A schematic diagram of converting a similar image block to a frequency domain according to an exemplary embodiment of the present disclosure is shown. Referring to Figure 6 The first image is divided into 8x8 initial image blocks, and the similar image blocks of the initial image blocks are 8x8 pixel blocks. Each similar image block includes 64 pixel points. After DCT (Discrete Cosine Transform) conversion is performed on each 8x8 similar image block, a corresponding 8x8 frequency domain image block is obtained, and each frequency domain image block includes 64 frequency components (such as various cosine components). That is, each similar image block corresponds to 64 frequency components.

[0050] Each initial image block of the first image is converted from the spatial domain to the frequency domain in the above conversion manner, so as to determine the frequency components of each similar image block.

[0051] It should be noted that the present disclosure is described by taking DCT conversion as an example, and a corresponding conversion manner can be selected according to the actual application scene to obtain the frequency components corresponding to each similar image block, and no special requirement is made.

[0052] In step S340, the frequency components of each similar image block are fused, and the result after the frequency component fusion is converted in the spatial domain to obtain a denoised image block corresponding to the initial image block.

[0053] In the exemplary embodiments of the present disclosure, for any initial image block, the frequency components of the similar image blocks in the similar image block set corresponding to the any initial image block are fused to obtain the fused frequency components.

[0054] Exemplarily, taking that the similar image block set of the initial image block C includes 2n+1 similar image blocks as an example, after frequency domain conversion is performed on each similar image block, 2n+1 frequency domain image blocks are obtained, and each frequency domain image block includes N frequency components. The fusion of the frequency components of the similar image blocks is the corresponding fusion of the frequency components of the 2n+1 similar image blocks for each frequency component. For example, the first frequency components of the 2n+1 similar image blocks are fused to obtain a first fused frequency component, the second frequency components of the 2n+1 similar image blocks are fused to obtain a second fused frequency component, and so on, to obtain an Nth fused frequency component. That is, after the fusion of the frequency components of the similar image blocks, N fused frequency components are obtained, and the frequency domain image block corresponding to the N fused frequency components is obtained.

[0055] The similar image blocks in the similar image block set of all initial image blocks of the first image are fused according to the above fusion manner, the results after the fusion of the frequency components are obtained respectively, and the results after the fusion of the frequencies are subjected to spatial domain conversion to obtain the denoised image blocks corresponding to the initial image blocks.

[0056] The fusion of the frequency components of the similar image blocks enables the denoised image blocks to increase the texture information corresponding to the high frequency components from the similar image blocks and the image flat area information corresponding to the low frequency components, and improves the performance of the denoised image blocks in the high frequency and low frequency aspects.

[0057] Step S350: generating a denoised image according to the denoised image blocks corresponding to the initial image blocks.

[0058] In the exemplary embodiments of the present disclosure, after the denoised image blocks corresponding to the initial image blocks are obtained respectively according to the steps S330 and S340, a denoised image can be generated according to the denoised image blocks corresponding to the initial image blocks.

[0059] Figure 7 A schematic diagram of one embodiment of an image processing method is shown, and the image set includes a dark frame image and a bright frame image. The brightness of the bright frame image is greater than a brightness threshold, the brightness of the dark frame image is less than or equal to the brightness threshold, and the dark frame image is used as a to-be-denoised image.

[0060] Firstly, the dark frame image is divided into a plurality of initial image blocks, each initial image block is a 8x8 pixel block, the similar image blocks of the initial image blocks can be obtained from the dark frame image, and the similar image blocks of the initial image blocks can also be obtained from the bright frame image to form a similar image block set of the initial image blocks; or, the similar image block set of the initial image blocks can be obtained only from the bright frame image.

[0061] Secondly, frequency domain conversion is performed on each similar image block in the similar image block set to determine the frequency components of the similar image blocks.

[0062] Then, the frequency components of each similar image block are fused to obtain a result of frequency component fusion, and finally 64 frequency components are obtained after fusion.

[0063] Further, the result of frequency component fusion is subjected to spatial domain conversion to obtain a noise-reduced image block corresponding to the initial image block.

[0064] Finally, after all initial image blocks obtain corresponding noise-reduced image blocks, each noise-reduced image block is fused to generate a noise-reduced image, thereby completing the noise reduction processing of the dark frame image.

[0065] By combining the bright frame image to obtain the similar image block set of the initial image block of the dark frame image, the number of reference image blocks and the image brightness information in the processing process are increased, and since the noise intensity of the bright frame image is small, more real image information can be retained, thereby improving the texture restoration of the noise-reduced image. In addition, by fusing the frequency components of each similar image block, a balance between high-frequency information and low-frequency information of the image can be sought, avoiding the case that the high-frequency information and the low-frequency information of the noise-reduced image are too high or too low, retaining the image texture information while making the image smooth transition.

[0066] In an exemplary embodiment, an implementation of determining a similar image block set corresponding to an initial image block is provided. Determining the similar image block set corresponding to the initial image block from the image set can include steps S810 to S830:

[0067] Step S810: obtaining a first similar image block corresponding to the initial image block from the first image.

[0068] The way of obtaining the first similar image block of the initial image block in the first image can be according to the obtaining method described in step S320, which will not be repeated here.

[0069] Step S820: obtaining a second similar image block corresponding to the initial image block from the second image.

[0070] In addition to obtaining the similar image block of the initial image block in the first image, the second similar image block of the initial image block is also obtained from the second image.

[0071] In some possible embodiments, since the second image and the first image are images with different brightness of the same shooting scene, the second similar image block corresponding to the position of the first similar image block in the first image can be obtained in the second image according to the position of the first similar image block, and the image content of the second similar image block is basically aligned with that of the first similar image block.

[0072] In some possible embodiments, a matching image block can also be obtained in the second image first, and then a corresponding image block of the matching image block can be obtained in the second image block as the second similar image block. By searching for the similar image blocks of the initial image block and the matching image block in the first image and the second image respectively, the search range of the similar image blocks can be further increased.

[0073] Since the first image and the second image of different brightness carry different image information, the number of reference image blocks in the noise reduction processing can be increased and the image reference information can be enriched according to the similar image blocks from the images of different brightness.

[0074] Step S830: determining a set of similar image blocks corresponding to the initial image block according to the first similar image block and the second similar image block.

[0075] After the first similar image block and the second similar image block are obtained, a set of similar image blocks corresponding to the initial image block is determined according to the obtained first similar image block and the second similar image block.

[0076] It should be noted that any initial image block can obtain a corresponding set of similar image blocks according to steps S810 to S830. In some possible embodiments, since the process of searching for the first similar image block is performed in the whole domain of the first image, the initial image block can also belong to its own similar image block, that is, the set of similar image blocks can include the initial image block corresponding to the set of similar image blocks.

[0077] In an example embodiment, in order to perform noise reduction processing on the first image, another implementation manner of determining a set of similar image blocks corresponding to the initial image block is provided. Determining a set of similar image blocks corresponding to each initial image block from the image set can obtain a set of similar image blocks corresponding to each initial image block in the second image. For example, a set of similar image blocks corresponding to an initial image block in a dark frame image can be obtained from a bright frame image, so that the noise reduction of the dark frame image is realized by using the bright frame image with smaller noise intensity. Correspondingly, the bright frame image can also be reduced by using the dark frame image, which realizes the effect of combining the image information of images of different brightness and reduces the calculation amount to a certain extent. The set of similar images can also include the initial image block itself.

[0078] In an example embodiment, the initial image block in the set of similar image blocks is taken as the center, and the frequency reduction is realized by the difference of the frequency components of the initial image block and other similar image blocks in the set of similar image blocks. The frequency components of each similar image block are fused, and the spatial domain conversion is performed on the result after the fusion of the frequency components, so as to obtain a noise-reduced image block corresponding to the initial image block, which can include steps S910 to S940:

[0079] Step S910: determining frequency difference information between the frequency components of each similar image block and the frequency components of the initial image block corresponding to the similar image block set.

[0080] The frequency difference information refers to the frequency component difference value between the similar image blocks in the similar image block set and the initial image block in the similar image block set.

[0081] For example, if the similar image block set includes 2n+1 (0, 1, 2…2n) similar image blocks (8x8 pixel blocks), each of which includes N frequency components. The 0th similar image block is the initial image block itself. When fusing the first frequency components, the difference value between the first frequency component of the 1st similar image block and the first frequency component of the 0th similar image block can be determined, the difference value between the first frequency component of the 2nd similar image block and the first frequency component of the 0th similar image block can be determined, and so on, until the difference value between the first frequency component of the 2nth similar image block and the first frequency component of the 0th similar image block is obtained. The process of fusing the second frequency component to the Nth frequency component is similar, which will not be listed one by one here.

[0082] The frequency difference value between the frequency components of the similar image blocks and the frequency components of the initial image block corresponding to the similar image block set is used to determine the weight of the frequency component fusion according to the frequency difference value, wherein the greater the frequency difference value, the smaller the weight corresponding to the frequency component of the similar image block, and vice versa.

[0083] Step S920: determining the weight corresponding to the frequency component of each similar image block according to the frequency difference information corresponding to the frequency component of each similar image block and the noise intensity of the first image.

[0084] The embodiment of the present disclosure can adjust the weight of the frequency component of the similar image block according to the noise intensity of the first image, determine the weight corresponding to the frequency component of each similar image block in combination with the frequency difference information and the noise intensity of the first image, and consider the difference of the frequency components while being conducive to maintaining the definition of the image after noise reduction.

[0085] For example, the weight corresponding to the frequency component of each similar image block can be determined by the following formula (1):

[0086]

[0087] Wherein, A Z (W) is the weight of the Wth frequency component of the Zth similar image block, F0(W) is the Wth frequency component of the 0th similar image block (initial image block) in the similar image block set, F ZF0(W) is the Wth frequency component of the Zth similar image block in the similar image block set, and ε is a noise intensity of the first image. 2 is a variance of the noise intensity of the first image, d is a variance adjustment factor, and is used to adjust the weight value according to actual needs.

[0088] Step S930: The frequency components of the similar image blocks are fused by weighting based on the weights corresponding to the frequency components of the similar image blocks, to obtain a result of frequency component fusion.

[0089] After the weights corresponding to the frequency components of the similar image blocks are obtained, the frequency components of the similar image blocks are fused by weighting according to the weights corresponding to the frequency components of the similar image blocks. For example, a result of frequency component fusion can be obtained by using formula (2).

[0090]

[0091] F0(W) is the Wth frequency component of the Zth similar image block in the similar image block set, and ε is a noise intensity of the first image. Z F0(W) is the Wth frequency component of the Zth similar image block in the similar image block set, and ε is a noise intensity of the first image.

[0092] In addition, the result of the Wth frequency component fusion can also be obtained by using an average value method, as shown in formula (3):

[0093]

[0094] F0(W) is the Wth frequency component of the Zth similar image block in the similar image block set, and ε is a noise intensity of the first image. Z F0(W) is the Wth frequency component of the Zth similar image block in the similar image block set, and ε is a noise intensity of the first image.

[0095] The result of frequency component fusion is obtained for each frequency component of each similar image block according to steps S910 to S930, and a frequency domain image block after noise reduction in a frequency domain of the initial image block can be obtained according to the obtained results of frequency component fusion.

[0096] Step S940: The results of frequency component fusion are converted in a spatial domain to obtain a noise reduction image block corresponding to the initial image block.

[0097] The result of frequency component fusion is converted in a spatial domain from a frequency domain to a spatial domain to obtain a noise reduction image block corresponding to the initial image block.

[0098] It should be noted that the similar image block set corresponding to each initial image block in the first image performs steps S910 to S940 to obtain a noise reduction image block corresponding to each initial image block.

[0099] Since the weight of each frequency component is determined respectively when the frequency components are fused, rather than using the same fusion coefficient in the whole frequency domain processing process, the problem of image blurring after noise reduction is avoided; and the variance of the noise intensity of the first image is introduced in the process of determining the weight of the frequency component, so that the weight can be adjusted according to the variance, and then the noise reduction intensity of the frequency component is adjusted to maintain the clarity of the image after noise reduction.

[0100] In an exemplary embodiment, in order to improve the efficiency of frequency component fusion, another implementation manner for determining the weight corresponding to the frequency component of each similar image block is also provided. The weight corresponding to the frequency component of each similar image block can also include:

[0101] The frequency components of the similar image block are first grouped, and then the target weight is determined from the weight corresponding to the frequency component in each group, wherein the target weight is determined as the weight corresponding to the frequency component in the corresponding group.

[0102] For example, if the similar image block includes 64 frequency components, the 64 frequency components are divided into 8 groups (1st group, 2nd group…8th group) with 8 frequency components as a group. A target weight is determined from the weight corresponding to the frequency component in the 1st group, and the target weight is determined as the weight corresponding to each frequency component in the 1st group. Similarly, the target weights corresponding to other groups are obtained respectively, that is, the target weight is the weight when the frequency components in the corresponding group are fused.

[0103] In some possible embodiments, the weight corresponding to a frequency component in a group can be randomly selected as the target weight, or the average value of the weights corresponding to the frequency components in a group can be taken as the target weight of the group, and no special limitation is made.

[0104] By using the same weight for multiple frequency components, the problem of image blurring after noise reduction caused by using the same fusion coefficient in the whole frequency fusion process is avoided, and the fusion processing efficiency is improved, and the noise reduction efficiency is improved.

[0105] In an exemplary embodiment, the second image can also be subjected to brightness alignment processing based on the first image. Brightness alignment is a processing process of adjusting the brightness distribution of the second image to be consistent with the brightness distribution of the first image.

[0106] The ratio between the exposure time length of the first image and the exposure time length of the second image can be obtained, and the brightness of the second image is adjusted based on the ratio to obtain the second image with the brightness aligned with the first image. Alternatively, if the first image and the second image are images with different brightness obtained under the same exposure time length, such as the bright-dark images output by the DCG, the ratio can be determined according to the proportional relationship of the gain values between the bright-dark images. Of course, the brightness alignment of the second image with the first image can also be other ways, and the embodiments of the present disclosure include but are not limited to the above brightness alignment method.

[0107] Since the second image and the first image are images with different brightness of the same shooting scene, the image textures of the two are basically similar, and the brightness is aligned to further improve the accuracy of the similar image block determination.

[0108] In an exemplary embodiment, in order to avoid the influence of the first image and the second image due to the large difference in noise intensity affecting the noise reduction effect, the noise intensity of the second image can be limited to the same level as the first image.

[0109] The noise intensity ratio between the first image and the second image can be obtained, the target noise intensity corresponding to the noise intensity ratio is determined, and the pre-noise reduction intensity of the first image and the second image is determined respectively, and the first image is pre-noise reduction processed according to the pre-noise reduction intensity corresponding to the first image, and the second image is pre-noise reduction processed based on the pre-noise reduction intensity corresponding to the second image.

[0110] The target noise intensity is obtained according to the preset noise intensity ratio and noise intensity corresponding relationship, wherein the preset noise intensity ratio and noise intensity corresponding relationship can be determined before the target noise intensity corresponding to the noise intensity ratio is determined to determine the pre-noise reduction intensity of the first image and the second image.

[0111] In an exemplary embodiment, the preset noise intensity ratio and noise intensity corresponding relationship can be determined by the noise intensity ratio of the reference images with different brightness under different exposure time lengths. Wherein, the first reference image and the second reference image of the same shooting scene can be collected, the brightness of the first reference image is greater than the brightness threshold, the brightness of the second reference image is less than or equal to the brightness threshold, and the preset noise intensity ratio and noise intensity corresponding relationship is determined according to the noise intensity ratio of the first reference image and the second reference image corresponding to different exposure time lengths.

[0112] The same shooting scene includes a shooting scene under the same lighting condition, such as shooting the same shooting scene under the same light source. For example, Figure 10As shown, first, two reference images (a first reference image and a second reference image) with different brightness are acquired simultaneously at different exposure times to obtain pairs of light and dark images corresponding to different exposure times. The noise intensity of each image is counted to obtain the ratio N1 / N2 of the noise intensity N1 of the first reference image to the noise intensity N2 of the second reference image. Finally, a curve of the ratio N0 / N1 varying with exposure time is fitted. Since exposure time is related to noise intensity, the corresponding relationship between the noise intensity ratio and noise intensity is obtained.

[0113] By fitting the correspondence between the noise intensity ratio and the noise intensity of two reference images with different brightness, during actual noise reduction processing, the noise reduction intensity required for the light and dark images can be obtained respectively based on the correspondence between the noise intensity ratio and the noise intensity, thereby limiting the noise levels of the first image and the second image to the same level according to the respective required noise reduction intensities, thereby improving the pre-noise reduction efficiency.

[0114] In an exemplary embodiment, in order to ensure the accuracy of similar image blocks of the initial image blocks, the image content of the second image can be aligned based on the first image before the first image is divided into multiple initial image blocks and the similar image block set corresponding to each initial image block is determined from the image set. Although in some application scenarios, such as using DCG technology to simultaneously obtain light and dark images with the same exposure time, the image content of the two frames is basically aligned. However, in some application scenarios, although the images are obtained from the same shooting scene, there may be differences in the image content. For example, the content difference between the first image and the second image may be caused by the presence of dynamically changing shooting objects (such as people, animals, etc.) in the shooting scene during the shooting process or the shaking of the terminal device.

[0115] Image content alignment involves aligning the texture and other content of the second image with the first image to improve the accuracy of the acquired similar image blocks. Image content alignment can be achieved through processes such as key point detection, key point matching, and image transformation, including but not limited to algorithms such as scale-invariant feature transformation, and is not specifically limited to these methods.

[0116] In an exemplary embodiment, the first image may also be obtained by fusing at least two images with different brightness in the image set to be processed.

[0117] Among them, the first image can be obtained by fusing the bright frame image and the dark frame image in the image set, and then the method of the embodiment of the present invention is used to further reduce the noise of the image after the fusion of the bright frame image and the dark frame image, and the image after the light and dark fusion is again reduced in noise using images of different brightness, thereby increasing the performance of the image after noise reduction in image brightness, which further helps to achieve the effect of high dynamic range imaging.

[0118] Figure 11A flowchart of an image processing method in an application scenario to which an example embodiment of the present disclosure relates is shown as follows. Figure 11 The image processing method of the present embodiment is described below with the dark frame image as the first image (the image to be denoised) as an example.

[0119] First, the first image and the second image are pre-processed by referring to the preset noise intensity ratio and noise intensity correspondence, so that the first image and the second image are at the same noise level.

[0120] Second, the second image is processed for brightness alignment with the first image (dark frame) as a reference.

[0121] Then, the second image is processed for image content alignment based on the first image.

[0122] The above process realizes the pre-processing of the first image and the second image.

[0123] Further, the dark frame image is processed for denoising to obtain a denoised image.

[0124] The first image is divided into a plurality of initial image blocks, and a set of similar image blocks corresponding to each initial image block is determined from the image set, and a denoised image block corresponding to each initial image block is obtained based on the set of similar image blocks.

[0125] The denoised image is generated according to the denoised image blocks corresponding to each initial image block.

[0126] The method for obtaining the denoised image block corresponding to any initial image block includes:

[0127] The frequency components of each similar image block in the set of similar image blocks are determined by frequency domain conversion.

[0128] The frequency components of each similar image block are fused, and the result after frequency component fusion is subjected to spatial domain conversion to obtain the denoised image block corresponding to the initial image block.

[0129] The method for obtaining the similar image block of the initial image block and the method for obtaining the weight corresponding to each frequency component have been described above and will not be repeated here.

[0130] It can be seen from the above that the similar image block set corresponding to the initial image block is increased with image blocks of different brightness, which increases the reference information of the noise reduction processing and enriches the image brightness information at each frequency component fusion stage, which is helpful to realize high dynamic range imaging; the frequency components of each similar image block are fused, so that the noise reduction image block is increased with the texture information corresponding to the high frequency components from each similar image block and the image flat area information corresponding to the low frequency components, which improves the performance of the noise reduction image block in high frequency and low frequency, so that the noise reduction image block retains the image edge and texture information while realizing the effect of smoothing the image, which is beneficial to improve the noise reduction effect of the image.

[0131] It should be noted that the above figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not for limiting purposes. It is easy to understand that the processes shown in the above figures do not indicate or limit the time sequence of the processes. In addition, it is also easy to understand that these processes can be executed synchronously or asynchronously, for example, in multiple modules.

[0132] Further, referring to Figure 12 It is shown that an image processing device 1200 is provided in the exemplary embodiments of the present disclosure, which includes an acquisition module 1210, a first processing module 1220, a second processing module 1230 and a third processing module 1240. Wherein:

[0133] The acquisition module 1210 is configured to acquire an image set, the image set including a first image and at least one second image, the first image and the second image being images of different brightness of the same shooting scene;

[0134] The first processing module 1220 is configured to divide the first image into a plurality of initial image blocks, and determine a similar image block set corresponding to each initial image block from the image set;

[0135] The image conversion module 1230 is configured to perform frequency domain conversion on each similar image block in the similar image block set to determine the frequency components of each similar image block;

[0136] The second processing module 1240 is configured to fuse the frequency components of each similar image block, and perform spatial domain conversion on the result after the frequency component fusion to obtain a noise reduction image block corresponding to the initial image block;

[0137] The third processing module 1250 is configured to generate a noise reduction image according to the noise reduction image block corresponding to each initial image.

[0138] In an example embodiment, the first processing module 1220 is configured to: obtain a first similar image block corresponding to the initial image block from the first image; obtain a second similar image block corresponding to the initial image block from the second image; and determine the set of similar image blocks corresponding to the initial image block according to the first similar image block and the second similar image block.

[0139] In an example embodiment, the first processing module 1220 is configured to: obtain a second similar image block corresponding to the position of the first similar image block in the second image based on the position of the first similar image block in the first image.

[0140] In an example embodiment, the first processing module 1220 is configured to: determine a matching image block in the second image that matches the position of the initial image block; and obtain a second similar image block corresponding to the matching image block in the second image.

[0141] In an example embodiment, the first processing module 1220 is configured to: obtain the set of similar image blocks corresponding to each initial image block in the second image.

[0142] In an example embodiment, the set of similar image blocks includes the initial image block corresponding to the set of similar image blocks; the second processing module 1240 is configured to: determine frequency difference information between a frequency component of each similar image block and a frequency component of the initial image block corresponding to the set of similar image blocks; determine a weight corresponding to the frequency component of each similar image block according to the frequency difference information corresponding to the frequency component of each similar image block and a noise intensity of the first image; and perform weighted fusion on the frequency components of each similar image block based on the weight corresponding to the frequency component of each similar image block to obtain a result of frequency component fusion; and perform spatial domain conversion on each result of frequency component fusion to obtain a denoised image block corresponding to the initial image block.

[0143] In an example embodiment, the second processing module 1240 is configured to: group the frequency components of the similar image blocks; and determine a target weight corresponding to each group from the weights corresponding to the frequency components in the group, wherein the target weight is determined as the weight corresponding to the frequency components in the corresponding group.

[0144] In an example embodiment, the image processing apparatus 1200 further includes:

[0145] The brightness processing module is configured to: obtain a ratio between an exposure time of the first image and an exposure time of the second image; and adjust the brightness of the second image based on the ratio to obtain the second image with brightness aligned with the first image.

[0146] In an example embodiment, the image processing apparatus 1200 further includes:

[0147] The noise reduction module is configured to: acquire a noise intensity ratio between the first image and the second image; determine a pre-noise reduction intensity of the first image and the second image respectively according to a target noise intensity corresponding to the noise intensity ratio, the target noise intensity being obtained according to a preset corresponding relationship between noise intensity ratios and noise intensities; and perform pre-noise reduction processing on the first image according to the pre-noise reduction intensity corresponding to the first image, and perform pre-noise reduction processing on the second image based on the pre-noise reduction intensity corresponding to the second image.

[0148] In an example embodiment, the noise reduction module is further configured to:

[0149] acquire a first reference image and a second reference image of a same shooting scene, the first reference image having a brightness greater than a brightness threshold, and the second reference image having a brightness less than or equal to the brightness threshold; and determine the preset corresponding relationship between noise intensity ratios and noise intensities according to a corresponding relationship between noise intensity ratios of the first reference image and the second reference image and different exposure time lengths.

[0150] In an example embodiment, the image processing apparatus 1200 further includes:

[0151] The image content processing module is configured to perform alignment processing on the second image based on the first image in terms of image content.

[0152] In an example embodiment, the first image is obtained by fusing at least two images with different brightness in the image set.

[0153] The specific details of each module in the above apparatus have been described in detail in the method embodiments, and the undisclosed details can be referred to the embodiment contents of the method part, and thus will not be described again.

[0154] Those skilled in the art can understand that each aspect of the present disclosure can be implemented as a system, a method or a program product. Therefore, each aspect of the present disclosure can be embodied as a whole hardware embodiment, a whole software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuitry", "module" or "system".

[0155] In the example embodiments of the present disclosure, an electronic device for the above method is also provided, which can be the above image device or server. Generally, the electronic device at least includes a processor and a memory, the memory is used to store executable instructions of the processor, and the processor is configured to execute the above method by executing the executable instructions.

[0156] The following describes the above method in detail. Figure 13Taking the mobile terminal 1300 in the embodiment as an example, the structure of the electronic device in the embodiment of the present disclosure is exemplarily described. It should be understood by those skilled in the art that in addition to the components specifically used for mobile purposes, Figure 13 The structure in FIG. 1 can also be applied to fixed type devices. In other embodiments, the mobile terminal 1300 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure can be implemented in hardware, software, or a combination of software and hardware. The interface connection relationship between the components is only shown schematically and does not constitute a structural limitation of the mobile terminal 1300. In other embodiments, the mobile terminal may also adopt the same Figure 13 Different interface connection methods, or a combination of multiple interface connection methods.

[0157] like Figure 13 As shown, the mobile terminal 1300 may specifically include: a processor 1301, a memory 1302, a bus 1303, a mobile communication module 1304, an antenna 1, a wireless communication module 1305, an antenna 2, a display screen 1306, a camera module 1307, an audio module 1308, a power module 1309, and a sensor module 1310.

[0158] The processor 1301 may include one or more processing units, for example: the processor 1301 may include an AP (Application Processor), a modem processor, a GPU (Graphics Processing Unit), an ISP (Image Signal Processor), a controller, an encoder, a decoder, a DSP (Digital Signal Processor), a baseband processor and / or an NPU (Neural-Network Processing Unit), etc.

[0159] An encoder can encode (i.e., compress) an image or video to reduce data size for storage or transmission. A decoder can decode (i.e., decompress) encoded data of an image or video to restore the image or video data. The mobile terminal 1300 can support one or more encoders and decoders, such as JPEG (Joint Photographic Experts Group), PNG (Portable Network Graphics), BMP (Bitmap), and the like for image formats, and MPEG (Moving Picture Experts Group) 1, MPEG 10, H.1063, H.1064, HEVC (High Efficiency Video Coding), and the like for video formats.

[0160] The processor 1301 can form a connection with the memory 1302 or other components through the bus 1303.

[0161] The memory 1302 can be used to store computer-executable program codes including instructions. The processor 1301 executes various functional applications and data processing of the mobile terminal 1300 by running the instructions stored in the memory 1302. The memory 1302 can also store application data, such as storing image, video, and the like.

[0162] The communication function of the mobile terminal 1300 can be implemented through the mobile communication module 1304, the antenna 1, the wireless communication module 1305, the antenna 2, the modem processor, and the baseband processor, and the like. The antenna 1 and the antenna 2 are used to transmit and receive electromagnetic wave signals. The mobile communication module 1304 can provide a mobile communication solution applied to the mobile terminal 1300, such as 3G, 4G, 5G, and the like. The wireless communication module 1305 can provide a wireless communication solution applied to the mobile terminal 1300, such as wireless local area network, Bluetooth, near field communication, and the like.

[0163] The display screen 1306 is used to implement display functions, such as displaying a user interface, an image, a video, and the like, and displaying an abnormal prompt information. The camera module 1307 is used to implement a shooting function, such as shooting an image, a video, and the like, to collect scene images. The audio module 1308 is used to implement an audio function, such as playing an audio, collecting a voice, and the like. The power module 1309 is used to implement a power management function, such as charging a battery, supplying power to the device, monitoring a battery state, and the like. The sensor module 1310 can include one or more sensors to implement corresponding sensing and detecting functions.

[0164] In addition, the exemplary embodiments of the present disclosure also provide a computer readable storage medium, which stores a program product capable of implementing the method described above. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program codes for causing the terminal device to perform the steps described in the "Exemplary Method" section above according to various exemplary embodiments of the present disclosure when the program product is run on the terminal device.

[0165] It should be noted that the computer readable medium shown in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0166] In the present disclosure, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present disclosure, the computer readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer readable program codes. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium, which can send, propagate or transmit the program for use by or in conjunction with an instruction execution system, device or apparatus. The program codes contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0167] Furthermore, the programming code can execute entirely on the user's computing device, partly on the user's computing device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider (ISP). It will be appreciated that other embodiments can comprise an electronic select map display system, such as those described herein, including an electronic select map display system which functions as described herein, but is not specifically disclosed herein.

[0168] Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the features disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the disclosure being indicated by the following claims.

Claims

1. An image processing method, characterized by, The method comprises the following steps: obtaining an image set, the image set comprising a first image and at least one second image, the first image and the second image being images of different brightness of the same shooting scene; dividing the first image into a plurality of initial image blocks, and determining a similar image block set corresponding to each initial image block from the image set; performing frequency domain conversion on each similar image block in the similar image block set to determine the frequency component of each similar image block; fusing the frequency components of each similar image block, and performing spatial domain conversion on the fused result of the frequency components to obtain a noise-reduced image block corresponding to the initial image block; generating a noise-reduced image according to the noise-reduced image blocks corresponding to each initial image block; wherein, before the step of dividing the first image into a plurality of initial image blocks, the method further comprises the following steps: obtaining a noise intensity ratio between the first image and the second image; determining the pre-noise reduction intensity of the first image and the second image respectively according to the target noise intensity corresponding to the noise intensity ratio, the target noise intensity being obtained according to a preset corresponding relationship between noise intensity ratio and noise intensity; performing pre-noise reduction processing on the first image and the second image respectively according to the pre-noise reduction intensity corresponding to each of the first image and the second image.

2. The method of claim 1, wherein, The step of determining a similar image block set corresponding to each initial image block from the image set comprises the following steps: obtaining a first similar image block corresponding to the initial image block from the first image; obtaining a second similar image block corresponding to the initial image block from the second image; determining the similar image block set corresponding to the initial image block according to the first similar image block and the second similar image block.

3. The method of claim 2, wherein, The step of obtaining a second similar image block corresponding to the initial image block from the second image comprises the following steps: based on the position of the first similar image block in the first image, obtaining the second similar image block corresponding to the position of the first similar image block in the second image.

4. The method of claim 2, wherein, The step of obtaining a second similar image block corresponding to the initial image block from the second image comprises the following steps: determining a matching image block matching the position of the initial image block in the second image; obtaining the second similar image block corresponding to the matching image block in the second image.

5. The method of claim 1, wherein, The step of determining a similar image block set corresponding to each initial image block from the image set comprises the following steps: obtaining a similar image block set corresponding to each initial image block in the second image.

6. The method of claim 1, wherein, The similar image block set comprises the initial image block corresponding to the similar image block set; the step of fusing the frequency components of each similar image block, and performing spatial domain conversion on the fused result of the frequency components to obtain a noise-reduced image block corresponding to the initial image block comprises the following steps: determining the frequency difference information between the frequency component of each similar image block and the frequency component of the initial image block corresponding to the similar image block set; determining the weight corresponding to the frequency component of each similar image block according to the frequency difference information corresponding to the frequency component of each similar image block and the noise intensity of the first image; weight corresponding to the frequency component of each of the similar image blocks based on the frequency difference information corresponding to the frequency component of each of the similar image blocks and the noise intensity of the first image, and performing spatial domain conversion on each of the frequency component fusion results to obtain a noise-reduced image block corresponding to the initial image block.

7. The method of claim 6, wherein, weight corresponding to the frequency component of each of the similar image blocks based on the frequency difference information corresponding to the frequency component of each of the similar image blocks and the noise intensity of the first image, and grouping the frequency components of the similar image blocks; determining a target weight from the weights corresponding to the frequency components in each group, wherein the target weight is determined as the weight corresponding to the frequency components in the corresponding group.

8. The method of claim 1, wherein, Before the first image is divided into a plurality of initial image blocks and the similar image block set corresponding to each of the initial image blocks is determined from the image set, the method further comprises: obtaining a ratio between the exposure time of the first image and the exposure time of the second image; adjusting the brightness of the second image based on the ratio to obtain a second image with brightness aligned with the first image.

9. The method of claim 1, wherein, Before the pre-noise-reduced intensity of the first image and the second image is respectively determined based on the target noise intensity corresponding to the noise intensity ratio, the method further comprises: obtaining a first reference image and a second reference image of the same shooting scene, the brightness of the first reference image being greater than a brightness threshold, and the brightness of the second reference image being less than or equal to the brightness threshold; determining the correspondence between the preset noise intensity ratio and the noise intensity according to the correspondence between the noise intensity ratio of the first reference image and the second reference image and different exposure times.

10. The method of claim 1, wherein, Before the first image is divided into a plurality of initial image blocks and the similar image block set corresponding to each of the initial image blocks is determined from the image set, the method further comprises: aligning the image content of the second image based on the first image.

11. The method according to any one of claims 1 to 10, characterized in that, The first image is obtained by fusing at least two images with different brightness in the image set.

12. An image processing apparatus characterized by comprising: comprises: an acquisition module configured to acquire an image set, the image set comprising a first image and at least one second image, the first image and the second image being images with different brightness of the same shooting scene; a first processing module configured to divide the first image into a plurality of initial image blocks and determine a similar image block set corresponding to each of the initial image blocks from the image set; an image conversion module configured to perform frequency domain conversion on each of the similar image blocks in the similar image block set to determine the frequency component of each of the similar image blocks; a second processing module configured to fuse the frequency components of each of the similar image blocks and perform spatial domain conversion on the frequency component fusion results to obtain a noise-reduced image block corresponding to the initial image block; a third processing module configured to generate a noise-reduced image based on the noise-reduced image blocks corresponding to each of the initial image blocks. The image processing device further comprises: a noise reduction module configured to perform: acquiring a noise intensity ratio between the first image and the second image; According to the target noise intensity corresponding to the noise intensity ratio, the pre-noise reduction intensity of the first image and the second image is determined respectively, and the target noise intensity is obtained according to a preset corresponding relationship between the noise intensity ratio and the noise intensity; and the first image and the second image are pre-noise reduction processed respectively according to the pre-noise reduction intensity corresponding to the first image and the second image respectively.

13. An electronic device, comprising: Comprise: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the method of any one of claims 1 to 11 via executing the executable instructions.

14. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1 to 11.

Images