Image denoising method, device, storage medium and electronic device

Abstract

The present disclosure relates to an image denoising method, device, storage medium and electronic device, belonging to the field of image processing, and capable of removing noise while retaining the texture details of the image very well. An image denoising method, comprising: obtaining an image array to be denoised, each pixel in the array only contains one color component, and there are center pixels and pixels of the same color component as the center pixel at alternate rows and / or alternate columns of the row and column where the center pixel of the image array to be denoised, and there are adjacent pixels and pixels of the same color component as the adjacent pixel at alternate rows and / or alternate columns of the row and column where the adjacent pixels adjacent to the center pixel are located; based on the difference between the center pixel and the center pixel and the difference between the adjacent pixels and the adjacent pixels, the local texture difference corresponding to each center pixel and pixel is determined; and denoising is performed based on the local texture difference and the center pixel and pixel is removed from the row and column.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing, and in particular, to an image denoising method, device, storage medium and electronic device. Background Art

[0002] In order to eliminate or reduce image noise, many denoising algorithms have been developed, including median filtering, edge adaptive filtering, bilateral filtering, block-matching and 3D filtering (BM3D), etc. However, these algorithms either remove a lot of image details while removing noise, or the algorithms are complex and not suitable for real-time image sensor circuit design. Summary of the invention

[0003] The purpose of the present disclosure is to provide an image denoising method, device, storage medium and electronic device, which can remove noise cleanly while retaining the texture details of the image very well, and the algorithm is simple and suitable for use in real-time image sensor circuit design.

[0004] According to a first embodiment of the present disclosure, there is provided an image denoising method, comprising: acquiring an image array to be denoised, wherein each pixel in the image array to be denoised contains only one color component, and for a central pixel of the image array to be denoised, there are central pixels with the same color component as the central pixel at alternate rows and / or alternate columns of the row and column where the central pixel is located, and for adjacent pixels adjacent to the central pixel, there are adjacent pixels with the same color component as the adjacent pixels at alternate rows and / or alternate columns of the row and column where the adjacent pixels are located; determining local texture differences corresponding to each of the central pixel's alternate rows and columns based on the difference between the central pixel and the central pixel's alternate rows and columns and the difference between the adjacent pixels and the adjacent pixels' alternate rows and columns; and performing denoising based on the local texture differences and the central pixel's alternate rows and columns.

[0005] Optionally, the image array to be denoised is a Bayer format image array, and is a 5×5 array.

[0006] Optionally, determining the local texture differences corresponding to the pixels in every other row and column of the central pixel based on the difference between the central pixel and the pixels in every other row and column of the central pixel and the difference between the adjacent pixels and the pixels in every other row and column of the adjacent pixels comprises:

[0007] Calculate the difference between the central pixel and the pixels in alternate rows and columns of the central pixel;

[0008] Calculate the difference between the adjacent pixel and the pixels in alternate rows and columns of the adjacent pixel;

[0009] When the direction of the adjacent pixels toward the adjacent pixels is parallel to and in the same direction as the direction of the central pixel toward the central pixel, the difference between the central pixel and the central pixel and the difference between the adjacent pixels and the adjacent pixels are added together and their absolute values ​​are taken to obtain the local texture differences corresponding to each of the central pixels.

[0010] Optionally, the denoising based on the local texture differences and the center pixel and every other row and column pixels includes: determining the weights corresponding to the respective local texture differences using a preset correspondence table or a preset functional relationship; and performing weighted processing on every other row and column pixels of the center pixel using the corresponding weights to obtain a denoising result.

[0011] Optionally, the greater the local texture difference is, the smaller the weight corresponding to the local texture difference is.

[0012] Optionally, the method further comprises: performing demosaic interpolation on the denoised image array to obtain a denoised image.

[0013] According to a second embodiment of the present disclosure, an image denoising device is provided, comprising: an acquisition module, used to acquire an image array to be denoised, wherein each pixel in the image array to be denoised contains only one color component, and for a central pixel of the image array to be denoised, there are central pixels with the same color component as the central pixel at alternate rows and / or alternate columns of the row and column where the central pixel is located, and for adjacent pixels adjacent to the central pixel, there are adjacent pixels with the same color component as the adjacent pixels at alternate rows and / or alternate columns of the row and column where the adjacent pixels are located; a local texture difference determination module, used to determine the local texture difference corresponding to each of the central pixel's alternate rows and columns based on the difference between the central pixel and the central pixel's alternate rows and columns and the difference between the adjacent pixels and the adjacent pixels' alternate rows and columns; and a denoising module, used to perform denoising based on the local texture difference and the central pixel's alternate rows and columns.

[0014] Optionally, the image array to be denoised is a Bayer format image array, and is a 5×5 array.

[0015] Optionally, the local texture difference determination module is also used to: calculate the difference between the center pixel and the center pixel's every other row and column pixels; calculate the difference between the adjacent pixel and the adjacent pixel's every other row and column pixels; add the difference between the center pixel and the center pixel's every other row and column pixels and the difference between the adjacent pixel and the adjacent pixel's every other row and column pixels, when the direction of the adjacent pixel toward the adjacent pixel is parallel and in the same direction as the direction of the central pixel toward the center pixel's every other row and column pixels, and take the absolute value to obtain the local texture difference corresponding to each center pixel's every other row and column pixels.

[0016] Optionally, the denoising module is further used to: determine the weights corresponding to each local texture difference using a preset correspondence table or a preset functional relationship; and perform weighted processing on the center pixel and every other row and column pixels using the corresponding weights to obtain a denoising result.

[0017] Optionally, the greater the local texture difference is, the smaller the weight corresponding to the local texture difference is.

[0018] Optionally, the device further comprises a mosaic module for performing mosaic interpolation on the denoised image array to obtain a denoised image.

[0019] According to a third embodiment of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the steps of the method according to the first embodiment of the present disclosure are implemented.

[0020] According to a fourth embodiment of the present disclosure, an electronic device is provided, including: a memory on which a computer program is stored; and a processor for executing the computer program in the memory to implement the steps of the method according to the first embodiment of the present disclosure.

[0021] By adopting the above technical solution, the local texture difference corresponding to each center pixel and every other row and column pixel is determined based on the difference between the center pixel and every other row and column pixel of the image array to be denoised, and the difference between the adjacent pixel and every other row and column pixel of the adjacent pixel, and then denoising is performed based on the local texture difference and the center pixel and every other row and column pixel. In this way, the center pixel and every other row and column pixel with the same color component as the center pixel can be selected for denoising, and the local texture difference can be used to estimate the contribution of each denoising point, that is, each center pixel and every other row and column pixel to the denoising result (for example, to estimate the appropriate weight value of each denoising point). Therefore, it not only has a simple algorithm and requires less logic resources and storage resources, and is very suitable for real-time image sensor design, but also can well retain the texture details of the image while removing the noise.

[0022] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0024] Figure 1 The figure is a flowchart of an image denoising method according to an embodiment of the present disclosure.

[0025] Figure 2 is a schematic diagram of an exemplary image array to be denoised.

[0026] Figure 3 An exemplary Bayer format image array schematic is shown.

[0027] Figure 4 A 5×5 Bayer format image array to be denoised is shown, wherein the central pixel is the B color component.

[0028] Figure 5 A 5×5 Bayer format image array to be denoised is shown, wherein the central pixel is the G color component.

[0029] Figure 6 An exemplary correspondence diagram of local texture differences and weights is shown.

[0030] Fig. 7A It is a schematic diagram of the edge adaptive filtering denoising effect according to the prior art.

[0031] Figure 7B Schematic diagram of the denoising effect of the image denoising method according to an embodiment of the present disclosure.

[0032] Figure 8 is a schematic block diagram of an image denoising device according to an embodiment of the present disclosure.

[0033] Fig. 9 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0034] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0035] Figure 1 FIG. 1 is a flow chart of an image denoising method according to an embodiment of the present disclosure. Figure 1 As shown, the method includes the following steps S11 to S13.

[0036] In step S11, an array of images to be denoised is obtained, wherein each pixel in the array of images to be denoised contains only one color component, and for a central pixel of the array of images to be denoised, there are central pixels having the same color component as the central pixel in alternate rows and / or alternate columns of the row and column where the central pixel is located, and for adjacent pixels adjacent to the central pixel, there are adjacent pixels having the same color component as the adjacent pixel in alternate rows and / or alternate columns of the row and column where the adjacent pixels are located.

[0037] Figure 2 is a schematic diagram of an exemplary image array to be denoised. Figure 2 The image array to be denoised is a 7×7 array, where B represents the color component, G represents the green component, and R represents the red component. The numbers after B, G, and R represent the row and column numbers of the pixel. For example, G32 represents a pixel of the green component, which is located in the 3rd row and the 2nd column. Figure 2 In the figure, the central pixel of the image array to be denoised is R44, the R components at the alternate rows of the central pixel R44 are R24 and R64, the R components at the alternate columns of the central pixel R44 are R42 and R46, the R pixels at both alternate rows and alternate columns of the central pixel R44 are R22, R26, R66, and R62, the adjacent pixels to the central pixel R44 are G34, G43, G54, and G45, and the G components at alternate rows, alternate columns, and both alternate rows and columns of the adjacent pixels G34, G43, G54, and G45 can be determined by referring to the determination method of the alternate rows and alternate columns of the central pixel R44. It should be understood by those skilled in the art that Figure 2 The image array to be denoised shown is only an example, and the present disclosure does not limit the number of arrays of the image array to be denoised and the arrangement of color components in each row and column.

[0038] The image array to be denoised may be an original image array that has not been processed in any way. For example, the CMOS image sensor may be configured with parameters to allow the CMOS image sensor to output an original image array that has not been processed in any way.

[0039] The image array to be denoised may be a Bayer format image array or an image array in another format, as long as it meets the array form requirements of the image array to be denoised described above. Figure 3 An exemplary Bayer format image array schematic diagram is shown. In addition, the number of arrays of the image array to be denoised can be any number. Figure 4 A 5×5 Bayer format image array to be denoised is shown, wherein the central pixel is the B color component. Figure 5The 5×5 Bayer format image array to be denoised is shown, and the central pixel thereof is the color component G. Of course, the central pixel of the 5×5 Bayer format image array to be denoised may also be R, which is not shown in the present disclosure.

[0040] In step S12, local texture differences corresponding to each of the center pixel's alternate rows and columns are determined based on the differences between the center pixel and alternate rows and columns of the center pixel and the differences between the adjacent pixels and alternate rows and columns of the adjacent pixels.

[0041] In one embodiment, the difference between the center pixel and the center pixel's alternate rows and columns, and the difference between the adjacent pixels and the adjacent pixels' alternate rows and columns can be calculated. Then, when the direction of the adjacent pixels toward the adjacent pixels is parallel to and in the same direction as the direction of the central pixel toward the center pixel's alternate rows and columns, the difference between the center pixel and the center pixel's alternate rows and columns and the difference between the adjacent pixels and the adjacent pixels' alternate rows and columns are added and the absolute value is taken to obtain the local texture difference corresponding to each central pixel's alternate rows and columns.

[0042] by Figure 4 Taking the Bayer format image array to be denoised as an example, the central pixel is B33, the central pixels of the central pixel B33 are B11, B13, B15, B31, B35, B51, B53 and B55 in alternate rows and columns, and the adjacent pixels of the central pixel B33 are G23, G32, G34 and G43. The adjacent pixels of the adjacent pixels of the adjacent pixels of the pixel G23 are G21, G25, G41 and G45 in alternate rows and columns, the adjacent pixels of the adjacent pixels of the adjacent pixels of the pixel G32 are G12, G52, G14 and G54 in alternate rows and columns, the adjacent pixels of the adjacent pixels of the adjacent pixels of the pixel G34 are G14, G54, G12 and G52 in alternate rows and columns, and the adjacent pixels of the adjacent pixels of the adjacent pixels of the pixel G43 are G41, G45, G21 and G25 in alternate rows and columns.

[0043] Then, since the direction of the center pixel B33 toward the center pixel every other row and column pixel B11, the direction of the adjacent pixel G43 toward the adjacent pixel every other row and column pixel G21, and the direction of the adjacent pixel G34 toward the adjacent pixel every other row and column pixel G12 are parallel and in the same direction, the local texture difference DIFF1 corresponding to the center pixel every other row and column pixel B11 can be obtained by using the absolute value of the sum of their differences, that is, DIFF1 = |(B33-B11)+(G43-G21)+(G34-G12)|. Similarly, the local texture differences corresponding to the other center pixels every other row and column pixels B13, B15, B31, B35, B51, B53 and B55 can be obtained, that is:

[0044] DIFF2=|(B33-B13)+(G32-G12)+(G34-G14)|

[0045] DIFF3=|(B33-B15)+(G32-G14)+(G43-G25)|

[0046] DIFF4=|(B33-B31)+(G23-G21)+(G43-G41)|

[0047] DIFF5=|(B33-B35)+(G23-G25)+(G43-G45)|

[0048] DIFF6=|(B33-B51)+(G23-G41)+(G34-G52)|

[0049] DIFF7=|(B33-B53)+(G32-G52)+(G34-G54)|

[0050] DIFF8=|(B33-B55)+(G32-G54)+(G23-G45)|

[0051] Similarly, for Figure 5 For the Bayer format denoised image array, since its central pixel is G, that is, G33, based on Figure 5 The Bayer format image array to be denoised is obtained, and the local texture differences corresponding to the center pixels G11, G13, G15, G31, G35, G51, G53 and G55 in alternate rows and columns are obtained as follows:

[0052] DIFF1=|(G33-G11)+(R43-R21)+(B34-B12)|

[0053] DIFF2=|(G33-G13)+(B32-B12)+(B34-B14)|

[0054] DIFF3=|(G33-G15)+(B32-B14)+(R43-R25)|

[0055] DIFF4=|(G33-G31)+(R23-R21)+(R43-R41)|

[0056] DIFF5=|(G33-G35)+(R23-R25)+(R43-R45)|

[0057] DIFF6=|(G33-G51)+(R23-R41)+(B34-B52)|

[0058] DIFF7=|(G33-G53)+(B32-B52)+(B34-B54)|

[0059] DIFF8=|(G33-G55)+(B32-B54)+(R23-R45)|

[0060] In addition, for a 5×5 Bayer format image array to be denoised, when its central pixel is R, the local texture difference obtained is similar to the local texture difference when the central pixel is B.

[0061] In step S13, denoising is performed based on the local texture difference and the center pixel and alternate row and column pixels.

[0062] In one embodiment, a preset correspondence table or a preset functional relationship can be used to determine the weights corresponding to each local texture difference, that is, to determine the appropriate weight value of each denoising point (that is, the center pixel and every other row and column pixel), and then the center pixel and every other row and column pixel are weightedly processed using the corresponding weights to obtain a denoising result.

[0063] by Figure 4 Taking the Bayer format denoised image array shown as an example, in step S13, it is necessary to first use a preset corresponding table or a preset functional relationship to obtain the weights WEI1, WEI2, WEI3, WEI4, WEI5, WEI6, WEI7, WEI8 corresponding to the local texture differences DIFF1, DIFF2, DIFF3, DIFF4, DIFF5, DIFF6, DIFF7, and DIFF8 obtained in step S12, and then use WEI1, WEI2, WEI3, WEI4, WEI5, WEI6, WEI7, and WEI8 to perform weighted processing on the center pixels B11, B13, B15, B31, B35, B51, B53, and B55 in alternate rows and columns, respectively, to obtain a denoising result.

[0064] One way to weight the treatment could be:

[0065] CENTER_B_DE=B11*WEI1+B13*WEI2+B15*WEI3+B31*WEI4+B35*WEI5+B51*WEI6+B53*WEI7+B55*WEI8, where CENTER_B_DE represents the denoising result.

[0066] Another way to weight the process could be:

[0067] CENTER_B_DE=(B11*WEI1+B13*WEI2+B15*WEI3+B31*WEI4+B35*WEI5+B51*WEI6+B53*WEI7+B55*WEI8) / (WEI1+WEI2+WEI3+WEI4+WEI5+WEI6+WEI7+WEI8).

[0068] for Figure 5 For the Bayer format image array to be denoised, the denoising process is the same as Figure 4 The denoising method of the Bayer format denoised image array shown is similar, that is, if the first weighting method is used, it can be obtained:

[0069] CENTER_G_DE=(G11*WEI1+G13*WEI2+G15*WEI3+G31*WEI4+G35*WEI5+G51*WEI6+G53*WEI7+G55*WEI8)

[0070] When the second weighting method is used, we can get:

[0071] CENTER_G_DE=(G11*WEI1+G13*WEI2+G15*WEI3+G31*WEI4+G35*WEI5+G51*WEI6+G53*WEI7+G55*WEI8) / (WEI1+WEI2+WEI3+WEI4+WEI5+WEI6+WEI7+WEI8).

[0072] For a 5×5 Bayer format image array to be denoised, when its central pixel is R, the denoising result is similar to that when the central pixel is B.

[0073] In one embodiment, the local texture difference and the weight satisfy the following corresponding relationship, that is, the larger the local texture difference, the smaller the weight corresponding to the local texture difference. The smaller the local texture difference, the smaller the difference from the central pixel, in which case the corresponding weight is larger; the larger the local texture difference, the larger the difference from the central pixel, in which case the corresponding weight is smaller.

[0074] Figure 6 An exemplary corresponding relationship diagram between local texture differences and weights is shown. Figure 6 As shown in the figure, the greater the local texture difference, the smaller the corresponding weight. Moreover, when the value of the local texture difference is small, the change of the weight is steeper. As the value of the local texture difference gradually increases, the change of the weight also gradually becomes gentler.

[0075] FIG7 is a comparison diagram of the denoising effect of the image denoising method according to an embodiment of the present disclosure and the prior art. In FIG7, the figure on the left is the denoising effect of the edge adaptive filtering according to the prior art, and the figure on the right is the denoising effect according to the embodiment of the present disclosure. It can be seen from the figure that the denoising effect according to the embodiment of the present disclosure is better than the denoising effect of the edge adaptive filtering.

[0076] By adopting the above technical solution, the local texture difference corresponding to each center pixel and every other row and column pixel is determined based on the difference between the center pixel and every other row and column pixel of the image array to be denoised, and the difference between the adjacent pixel and every other row and column pixel of the adjacent pixel, and then denoising is performed based on the local texture difference and the center pixel and every other row and column pixel. In this way, the center pixel and every other row and column pixel with the same color component as the center pixel can be selected for denoising, and the local texture difference can be used to estimate the contribution of each denoising point, that is, each center pixel and every other row and column pixel to the denoising result (for example, to estimate the appropriate weight value of each denoising point). Therefore, it not only has a simple algorithm and requires less logic resources and storage resources, and is very suitable for real-time image sensor design, but also can well retain the texture details of the image while removing the noise.

[0077] In one embodiment, the image denoising method according to the embodiment of the present disclosure may further include: performing demosaic interpolation on the denoised image array to obtain a denoised image. Common demosaicing methods include bilinear interpolation, cubic interpolation, etc., so that the image after denoising can be obtained.

[0078] Figure 8 FIG. 1 is a schematic block diagram of an image denoising device according to an embodiment of the present disclosure. Figure 8 As shown, the image denoising device includes: an acquisition module 81, used for acquiring an image array to be denoised, wherein each pixel in the image array to be denoised contains only one color component, and for a central pixel in the image array to be denoised, there are central pixels and pixels of the same color component as the central pixel in alternate rows and / or alternate columns of the row and column where the central pixel is located, and for adjacent pixels adjacent to the central pixel, there are adjacent pixels and pixels of the same color component as the adjacent pixels in alternate rows and / or alternate columns of the row and column where the adjacent pixels are located; a local texture difference determination module 82, used for determining the local texture difference corresponding to each central pixel and pixel of the same color component as the adjacent pixel in alternate rows and columns based on the difference between the central pixel and the central pixel and the difference between the adjacent pixel and the adjacent pixel and the adjacent pixel; a denoising module 83, used for denoising based on the local texture difference and the central pixel and pixel of the same color component as the adjacent pixel in alternate rows and columns.

[0079] By adopting the above technical solution, the local texture difference corresponding to each center pixel and every other row and column pixel is determined based on the difference between the center pixel and every other row and column pixel of the image array to be denoised, and the difference between the adjacent pixel and every other row and column pixel of the adjacent pixel, and then denoising is performed based on the local texture difference and the center pixel and every other row and column pixel. In this way, the center pixel and every other row and column pixel with the same color component as the center pixel can be selected for denoising, and the local texture difference can be used to estimate the contribution of each denoising point, that is, each center pixel and every other row and column pixel to the denoising result (for example, to estimate the appropriate weight value of each denoising point). Therefore, it not only has a simple algorithm and requires less logic resources and storage resources, and is very suitable for real-time image sensor design, but also can well retain the texture details of the image while removing the noise.

[0080] Optionally, the image array to be denoised is a Bayer format image array, and is a 5×5 array.

[0081] Optionally, the local texture difference determination module 82 is further used to: calculate the difference between the center pixel and the center pixel's every other row and column pixels; calculate the difference between the adjacent pixel and the adjacent pixel's every other row and column pixels; add the difference between the center pixel and the center pixel's every other row and column pixels and the difference between the adjacent pixel and the adjacent pixel's every other row and column pixels, when the direction of the adjacent pixel toward the adjacent pixel is parallel and in the same direction as the direction of the central pixel toward the center pixel's every other row and column pixels, and take the absolute value to obtain the local texture difference corresponding to each center pixel's every other row and column pixels.

[0082] Optionally, the denoising module 83 is further used to: determine the weights corresponding to each local texture difference using a preset corresponding table or a preset functional relationship; and perform weighted processing on the center pixel and every other row and column pixels using the corresponding weights to obtain a denoising result.

[0083] Optionally, the greater the local texture difference is, the smaller the weight corresponding to the local texture difference is.

[0084] Optionally, the device further comprises a mosaic module for performing mosaic interpolation on the denoised image array to obtain a denoised image.

[0085] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0086] Fig. 9 FIG. 7 is a block diagram of an electronic device 700 according to an exemplary embodiment. Fig. 9As shown, the electronic device 700 may include: a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input / output (I / O) interface 704, and a communication component 705.

[0087] The processor 701 is used to control the overall operation of the electronic device 700 to complete all or part of the steps in the above-mentioned image denoising method. The memory 702 is used to store various types of data to support the operation of the electronic device 700, and these data may include, for example, instructions for any application or method used to operate on the electronic device 700, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 703 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, which is used to receive external audio signals. The received audio signal may be further stored in the memory 702 or sent through the communication component 705. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules, and the above-mentioned other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 705 may include: Wi-Fi module, Bluetooth module, NFC module.

[0088] In an exemplary embodiment, the electronic device 700 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to perform the above-mentioned image denoising method.

[0089] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned image denoising method are implemented. For example, the computer-readable storage medium may be the above-mentioned memory 702 including program instructions, and the above-mentioned program instructions may be executed by the processor 701 of the electronic device 700 to complete the above-mentioned image denoising method.

[0090] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0091] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0092] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. An image denoising method, characterized in that: include: Acquire an image array to be denoised, wherein each pixel in the image array to be denoised contains only one color component, and for a central pixel of the image array to be denoised, there are central pixels with the same color component as the central pixel at alternate rows and / or alternate columns of the row and column where the central pixel is located, and for adjacent pixels adjacent to the central pixel, there are adjacent pixels with the same color component as the adjacent pixel at alternate rows and / or alternate columns of the row and column where the adjacent pixels are located; Calculate the difference between the central pixel and the pixels in alternate rows and columns of the central pixel; Calculate the difference between the adjacent pixel and the pixels in alternate rows and columns of the adjacent pixel; When the direction of the adjacent pixel toward the adjacent pixel is parallel to and in the same direction as the direction of the central pixel toward the central pixel, the difference between the central pixel and the central pixel and the difference between the adjacent pixel and the adjacent pixel are added together and their absolute values ​​are taken to obtain the local texture difference corresponding to each of the central pixels; Denoising is performed on every other row and column of pixels based on the local texture difference and the center pixel.

2. The method according to claim 1, characterized in that The image array to be denoised is a Bayer format image array, and is a 5×5 array.

3. The method according to claim 1, characterized in that The denoising is performed based on the local texture difference and the center pixel and alternate row and column pixels, comprising: Determine the weight corresponding to each of the local texture differences by using a preset correspondence table or a preset functional relationship; The corresponding weights are used to perform weighted processing on the center pixel and alternate row and column pixels to obtain a denoising result.

4. The method according to claim 3, characterized in that The greater the local texture difference is, the smaller the weight corresponding to the local texture difference is.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: The denoised image array is demosaiced and interpolated to obtain a denoised image.

6. An image denoising device, characterized in that: include: an acquisition module, for acquiring an image array to be denoised, wherein each pixel in the image array to be denoised contains only one color component, and for a central pixel of the image array to be denoised, there are central pixels of the same color component as the central pixel at alternate rows and / or alternate columns of the row and column where the central pixel is located, and for adjacent pixels adjacent to the central pixel, there are adjacent pixels of the same color component as the adjacent pixel at alternate rows and / or alternate columns of the row and column where the adjacent pixels are located; A local texture difference determination module, used for calculating the difference between the central pixel and the pixels in alternate rows and columns of the central pixel; Calculate the difference between the adjacent pixel and the pixels in alternate rows and columns of the adjacent pixel; When the direction of the adjacent pixel toward the adjacent pixel is parallel to and in the same direction as the direction of the central pixel toward the central pixel, the difference between the central pixel and the central pixel and the difference between the adjacent pixel and the adjacent pixel are added together and their absolute values ​​are taken to obtain the local texture difference corresponding to each of the central pixels; The denoising module is used for denoising based on the local texture difference and the center pixel and alternate row and column pixels.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 5 are implemented.

8. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 5.

Images