Local contrast enhancement method

Abstract

The invention discloses a local contrast enhancement method. The method comprises the following steps: inputting an original RGB three-channel image S; the method comprises the following steps: converting an original RGB three-channel image S into YUV data, and extracting a brightness channel Y channel as a grayscale image I; preprocessing the gray level image I to obtain a difference image delta; optimizing the difference image delta to obtain an optimized difference image delta 2; adding the gray level image I and the optimized difference image delta 2 to obtain a gray level image Ie after local contrast enhancement; replacing a Y channel in the YUV space with Ie, and converting YUV into RGB to obtain a color image after local contrast enhancement; by using the method, the permeability can be improved, and two technical problems that black and white edges are easy to generate and details are lost due to overexposure and darkness in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a local contrast enhancement method. Background Art

[0002] Local transparency enhancement (also called local contrast enhancement or brightness / contrast adjustment) is an important technique in image processing that aims to improve the visual quality of images, especially in low light, low contrast, or in cases where image details are difficult to discern. This technique makes the details in the image more prominent by adjusting the brightness and contrast of different areas in the image, thereby improving the transparency and look and feel of the image. Common techniques include histogram processing, AI enhancement, tone mapping, and other solutions.

[0003] USM (Unsharp Mask) local transparency enhancement is an image processing technology that enhances the edge contrast of an image by simulating the characteristics of human observation, that is, the human eye sees the so-called "Mach band" effect at the edge of a sharp brightness change, and is mainly used to improve the transparency and visual perception of an image, especially by enhancing the local contrast of the image. USM local transparency enhancement technology can effectively improve the transparency and clarity of an image, making the edges and details in the image more prominent. USM technology can not only be used for post-processing of images, but can also be embedded in the image processing system of a digital camera to improve the transparency and clarity of the captured image in real time with high efficiency.

[0004] However, over-sharpening may cause edge overshoot (Halo phenomenon), which may manifest to a certain extent as unnatural brightness changes near the edges, including effects that may look like black and white edges, and may cause overexposure or overdarkness resulting in loss of details. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the problems existing in the local permeability enhancement technology in the above-mentioned prior art, the present invention is proposed.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a local contrast enhancement method, which comprises the following steps:

[0008] Input original RGB three-channel image S;

[0009] Convert the original RGB three-channel image S to YUV data, and extract the brightness channel Y as the grayscale image I;

[0010] Preprocess the grayscale image I to obtain a differential image Δ;

[0011] Optimize the difference image Δ to obtain an optimized difference image Δ_2;

[0012] Add the grayscale image I and the optimized difference image Δ_2 to obtain the grayscale image I_e after local contrast enhancement;

[0013] Replace the Y channel in the YUV space with I_e, and convert YUV to RGB to obtain a color image with local contrast enhancement.

[0014] As a preferred solution of the local contrast enhancement method of the present invention, the preprocessing step is specifically as follows:

[0015] Perform Gaussian filtering on the grayscale image I to obtain a Gaussian filtered image I_GS, where the Gaussian filter radius is r;

[0016] The grayscale image I is used as the guide image, and the Gaussian filter image I_GS is subjected to a guide filter process to obtain a guide filter image I_GD, and the guide filter radius is r;

[0017] Calculate the difference image Δ, Δ = I-I_GD;

[0018] Here, r>1% of the minimum value of the image length and width.

[0019] As a preferred solution of the local contrast enhancement method of the present invention, the specific steps of optimizing the differential image Δ are:

[0020] Change the pixel values ​​in the differential image Δ whose absolute values ​​are not greater than the threshold value t to 0, and obtain the differential image Δ_0;

[0021] The differential image Δ_0 is enhanced, and the enhanced differential image is Δ_1, Δ_1 = Δ*λ;

[0022] Extract the portion of the difference image Δ_1 with a value less than 0 as the darkened image Δ_neg, and extract the portion of the difference image Δ_1 with a value greater than 0 as the brightened image Δ_pos;

[0023] Based on the darkening coefficient λ_1 and the grayscale image I, the darkened image Δ_neg is optimized, and the optimized darkened image is Δ_neg1;

[0024] Based on the brightening coefficient λ_2 and the grayscale image I, the brightened image Δ_pos is optimized, and the optimized brightened image is Δ_pos1;

[0025] Merge the updated Δ_neg and Δ_pos to obtain the optimized differential image Δ_2;

[0026] Among them, λ is the enhancement strength.

[0027] As a preferred solution of the local contrast enhancement method of the present invention, the step of optimizing the darkened image Δ_neg is:

[0028] Calculate the image darkening threshold t_neg_1=I*λ_1, 0<λ_1<1;

[0029] Calculate the dark area suppression coefficient λ_neg, λ_neg = y (i,j) 2 / (255+y (i,j) 2 ), where y (i,j) is the gray value corresponding to the position of the i-th row and j-th column of the image;

[0030] Calculate the dimming threshold t_neg_2 = -1*t_neg_1*λ_neg;

[0031] Δ_neg1=max(Δ_neg,t_neg_2).

[0032] As a preferred solution of the local contrast enhancement method of the present invention, the specific steps of optimizing and brightening the image Δ_pos are:

[0033] Calculate the image brightening threshold t_pos_1=(255-I)*λ_2, 0<λ_2<1;

[0034] Calculate the bright area suppression coefficient λ_pos, λ_neg = (255-y (i,j) ) 2 / (255+(255-y (i,j) ) 2 );

[0035] Calculate the brightening threshold t_pos_2 = t_pos_1*λ_pos;

[0036] Δ_pos1=max(Δ_pos,t_pos_2).

[0037] Beneficial effects of the present invention: Under the premise of improving transparency, the present invention solves two technical problems in the prior art, namely, easy generation of black and white edges and loss of details due to overexposure and darkness. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0039] Figure 1 It is a schematic diagram of the process of the present invention.

[0040] Figure 2 It is a curve diagram of dark area suppression coefficient and gray value in the present invention.

[0041] Figure 3 It is a curve diagram of the bright area suppression coefficient and the gray value in the present invention.

[0042] Figure 4 is a grayscale image I of an example.

[0043] Figure 5 Image I_e showing the black and white edges after enhancing an example using the common USM method.

[0044] Figure 6 This is the image I_e without black and white edges after being enhanced using the present invention.

[0045] Figure 7 is another example grayscale image I.

[0046] Figure 8 Another example of image I_e with detail loss after enhancement using the common USM method.

[0047] Fig. 9 FIG. 4 is another example of an image I_e without detail loss after being enhanced using the present invention. DETAILED DESCRIPTION

[0048] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0049] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0050] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0051] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0052] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0053] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0054] Example 1

[0055] Reference Figure 1 , which is the first embodiment of the present invention, and provides a local contrast enhancement method. This method is used to synthesize panoramic roaming videos without being associated with IP addresses. It can be directly downloaded and shared in video format, saving time and effort.

[0056] A local contrast enhancement method comprises the following steps:

[0057] S1: input original RGB three-channel image S;

[0058] S2: Convert the original RGB three-channel image S to YUV data and extract the brightness channel Y as the grayscale image I;

[0059] S3: preprocess the grayscale image I to obtain a differential image Δ;

[0060] The specific steps are:

[0061] S301 performs Gaussian filtering on the grayscale image I to obtain a Gaussian filtered image I_GS. The Gaussian filtering radius is r. The specific expression of the Gaussian kernel function GS is:

[0062]

[0063] I_GS = I*GS;

[0064] in is the Gaussian kernel variance, p and q are the relative coordinates within the Gaussian filter radius r;

[0065] S302 uses the grayscale image I as a guide image, performs guided filtering on the Gaussian filter image I_GS, and obtains a guided filter image I_GD. The guided filter radius is r. The specific guided filter function expression is as follows:

[0066]

[0067] where ω xy is a pixel block window with a radius of r centered at coordinate (x, y), ω ij is a pixel block window with a radius of r centered at coordinate (i, j), |ω xy | is the number of pixels in the pixel block window, a xy and b xy is the pixel block ω xy The window coefficient in I (i,j) is the pixel corresponding to the position of the i-th row and j-th column of the grayscale image, I (p,q) I_GS is the pixel corresponding to the pth row and qth column position of the grayscale image in the pixel block window. (p,q) is the pixel corresponding to the pth row and qth column position of the Gaussian filtered image in the pixel block window, μ is the guide image in the pixel block ω xy The mean value in To guide the image in pixel block ω xy The variance in , ε is the linear regression coefficient, is the image to be smoothed in pixel block ω xy The mean value in, p, q is the pixel block ω xy The relative coordinates within, i and j are image coordinates;

[0068] S303 calculates a differential image Δ, Δ=I-I_GD;

[0069] Where r>1% of the minimum value of the image length and width;

[0070] S4: Optimize the differential image Δ to obtain the optimized differential image Δ_2. The specific steps are:

[0071] S401, changing the pixel values ​​in the differential image Δ whose absolute values ​​are not greater than the threshold t to 0 to obtain the differential image Δ_0;

[0072] S402, enhancing the difference image Δ_0, the enhanced difference image is Δ_1, Δ_1 = Δ*λ, λ is the enhancement strength;

[0073] S403, extracting the portion of the difference image Δ_1 with a value less than 0 as a darkened image Δ_neg, and extracting the portion of the difference image Δ_1 with a value greater than 0 as a brightened image Δ_pos;

[0074] S404, based on the darkening coefficient λ_1 and the grayscale image I, the darkened image Δ_neg is optimized, and the optimized darkened image is Δ_neg1. The specific optimization steps are:

[0075] S404a, calculating the image darkening threshold t_neg_1=I*λ_1, 0<λ_1<1;

[0076] S404b, calculate the dark area suppression coefficient λ_neg, λ_neg = y (i,j) 2 / (255+y (i,j) 2 ), where y (i,j) is the gray value corresponding to the position of the i-th row and j-th column of the image;

[0077] S404c, calculating the dimming threshold t_neg_2=-1*t_neg_1*λ_neg;

[0078] S404d, Δ_neg1=max(Δ_neg,t_neg_2);

[0079] S405, based on the brightening coefficient λ_2 and the grayscale image I, optimize the brightened image Δ_pos, the optimized brightened image is Δ_pos1, and the optimization steps are:

[0080] S405a, calculating the image brightening threshold t_pos_1=(255-I)*λ_2, 0<λ_2<1;

[0081] S405b, calculate the bright area suppression coefficient λ_pos, λ_neg = (255-y (i,j) ) 2 / (255+(255-y (i,j) ) 2 ); S405c, calculate the brightening threshold t_pos_2 = t_pos_1*λ_pos;

[0082] S405d, Δ_pos1=max(Δ_pos,t_pos_2);

[0083] S406, merging the updated Δ_neg1 and Δ_pos1 to obtain an optimized differential image Δ_2, that is, Δ_2 = Δ_neg1 + Δ_pos1;

[0084] S5: Add the grayscale image I and the optimized differential image Δ_2 to obtain a grayscale image I_e after local contrast enhancement;

[0085] S6: Replace the Y channel in the YUV space with I_e, and convert YUV to RGB to obtain a color image with enhanced local contrast.

[0086] Among them, max(Δ_neg, t_neg_2) is the maximum value between Δ_neg and t_neg_2, and max(Δ_pos, t_pos_2) is the maximum value between Δ_pos and t_pos_2.

[0087] When the present invention is used to process an image, the two technical problems of the prior art that black and white edges are easily generated and details are lost due to overexposure and darkness are solved while improving the transparency.

[0088] Example 2

[0089] This embodiment provides a local contrast enhancement method, which is different from Embodiment 1 in that this embodiment compares the test results by means of scientific demonstration to verify the real effect of this method.

[0090] A local contrast enhancement method comprises the following steps:

[0091] S1: input original RGB three-channel image S;

[0092] S2: Convert the original RGB three-channel image S to YUV data and extract the brightness channel Y as the grayscale image I;

[0093] S3: preprocess the grayscale image I to obtain a differential image Δ;

[0094] The specific steps are:

[0095] S301 performs Gaussian filtering on the grayscale image I to obtain a Gaussian filtered image I_GS, where the Gaussian filtering radius is r;

[0096] S302 uses the grayscale image I as a guide image, performs guided filtering on the Gaussian filter image I_GS, and obtains a guided filter image I_GD, where the guided filter radius is r;

[0097] S303 calculates a differential image Δ, Δ=I-I_GD;

[0098] Where r>1% of the minimum value of the image length and width. For an image of 8000*4000, r is 50.

[0099] S4: Based on image s, image I, enhancement intensity λ and threshold information, the differential image Δ is optimized to solve the overexposure and dark problems caused by transition enhancement, and the optimized differential image Δ_2 is obtained. The specific steps are:

[0100] S401, changing the pixel values ​​in the differential image Δ whose absolute values ​​are not greater than the threshold value t_1 to 0, to obtain the differential image Δ_0. In this embodiment, the threshold value t is 4, which can be calibrated according to the actual situation;

[0101] S402, enhancing the difference image Δ_0, the enhanced difference image is Δ_1, Δ_1=Δ*λ, λ is the enhancement strength, which can be calibrated according to the actual situation;

[0102] S403, extracting the portion of the difference image Δ_1 with a value less than 0 as a darkened image Δ_neg, and extracting the portion of the difference image Δ_1 with a value greater than 0 as a brightened image Δ_pos;

[0103] S404, based on the darkening coefficient λ_1 and the grayscale image I, the darkened image Δ_neg is optimized, and the optimized darkened image is Δ_neg1. The optimization steps are:

[0104] S404a, calculating the image darkening threshold t_neg_1=I*λ_1, 0<λ_1<1, in this embodiment, λ_1 is 0.5;

[0105] S404b, calculate the dark area suppression coefficient λ_neg, λ_neg = y (i,j) 2 / (255+y (i,j) 2 )(like Figure 2 As shown), where y (i,j) is the gray value corresponding to the position of the i-th row and j-th column of the image;

[0106] S404c, calculating the dimming threshold t_neg_2=-1*t_neg_1*λ_neg;

[0107] S404d, Δ_neg1=max(Δ_neg,t_neg_2);

[0108] To improve calculation efficiency, the dark area suppression coefficient λ_neg corresponding to each gray value can be calculated and stored as a conversion table. When optimizing the darkened image Δ_neg, the corresponding dark area suppression coefficient λ_neg is directly extracted from the conversion table (as shown in the following table) according to the gray value for use;

[0109] Gray value Dark area suppression coefficient λ_neg 0 0 1 0.00390625 2 0.01544402 ... ... 254 0.99606305 255 0.99609375

[0110] S405, based on the brightening coefficient λ_2 and the grayscale image I, the brightened image Δ_pos is optimized, and the optimized brightened image is Δ_pos1. The specific optimization steps are:

[0111] S405a, calculating the image brightening threshold t_pos_1=(255-I)*λ_2, 0<λ_2<1, in this embodiment, λ_2 is 0.5;

[0112] S405b, calculate the bright area suppression coefficient λ_pos, λ_pos=(255-y (i,j) ) 2 / (255+(255-y (i,j) ) 2 )(like Figure 3 shown), y (i,j) is the gray value corresponding to the position of the i-th row and j-th column of the image;

[0113] S405c, calculating the brightening threshold t_pos_2=t_pos_1*λ_pos;

[0114] S405d, Δ_pos1=max(Δ_pos,t_pos_2);

[0115] To improve calculation efficiency, the bright area suppression coefficient λ_pos corresponding to each gray value can be calculated and stored as a conversion table. When optimizing the brightened image Δ_pos, the corresponding bright area suppression coefficient λ_pos is directly extracted from the conversion table (as shown in the following table) according to the gray value for use;

[0116]

[0117]

[0118] S406, merging the updated Δ_neg1 and Δ_pos1 to obtain an optimized differential image Δ_2;

[0119] S5: Add the grayscale image I and the optimized differential image Δ_2 to obtain a grayscale image I_e after local contrast enhancement;

[0120] S6: Replace the Y channel in the YUV space with I_e, and convert YUV to RGB to obtain a color image with enhanced local contrast.

[0121] like Figures 4 to 6 As shown, obtain an original RGB three-channel image, convert the original RGB three-channel image to YUV format, and extract the brightness channel Y channel as the grayscale image I, as shown Figure 4; The grayscale image I is enhanced by the common USM method. There is an obvious black and white edge phenomenon on the image, such as Figure 5 Through the technical solution of the present invention, the problem of black and white edges in the image is eliminated while the local contrast is enhanced.

[0122] like Figure 7 to Figure 9 As shown, obtain an original RGB three-channel image, convert the original three-channel image to YUV format, and extract the brightness channel Y channel as the grayscale image I, as shown Figure 7 ; When the grayscale image I is enhanced by the common USM method, the areas with high brightness on the image are overexposed, and the areas with low brightness are too dark, resulting in the loss of image details, such as Figure 8 Through the technical solution of the present invention, while performing local contrast enhancement processing, the problems of image overexposure and over-darkness are suppressed, and the loss of details is avoided.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the technical solutions of the present invention.

Claims

1. A local contrast enhancement method, characterized in that: The following steps are included: Input original RGB three-channel image S; Convert the original RGB three-channel image S to YUV data, and extract the brightness channel Y as the grayscale image I; Preprocess the grayscale image I to obtain a differential image Δ; Optimize the difference image Δ to obtain an optimized difference image Δ_2; Add the grayscale image I and the optimized difference image Δ_2 to obtain the grayscale image I_e after local contrast enhancement; Replace the Y channel in the YUV space with I_e, and convert YUV to RGB to obtain a color image with local contrast enhancement.

2. The local contrast enhancement method according to claim 1, characterized in that: The specific steps of preprocessing are: Perform Gaussian filtering on the grayscale image I to obtain a Gaussian filtered image I_GS, where the Gaussian filter radius is r; The grayscale image I is used as the guide image, and the Gaussian filter image I_GS is subjected to a guide filter process to obtain a guide filter image I_GD, and the guide filter radius is r; Calculate the difference image Δ, Δ = I-I_GD; Here, r>1% of the minimum value of the image length and width.

3. The local contrast enhancement method according to claim 2, characterized in that: The specific steps for optimizing the differential image Δ are: Change the pixel values ​​in the differential image Δ whose absolute values ​​are not greater than the threshold value t to 0, and obtain the differential image Δ_0; The differential image Δ_0 is enhanced, and the enhanced differential image is Δ_1, Δ_1 = Δ_0*λ; Extract the portion of the difference image Δ_1 with a value less than 0 as the darkened image Δ_neg, and extract the portion of the enhanced difference image Δ_1 with a value greater than 0 as the brightened image Δ_pos; Based on the darkening coefficient λ_1 and the grayscale image I, the darkened image Δ_neg is optimized, and the optimized darkened image is Δ_neg1; Based on the brightening coefficient λ_2 and the grayscale image I, the brightened image Δ_pos is optimized, and the optimized brightened image is Δ_pos1; Merge the updated Δ_neg and Δ_pos to obtain the optimized differential image Δ_2; Among them, λ is the enhancement strength.

4. The local contrast enhancement method according to claim 3, characterized in that: The steps to optimize the darkened image Δ_neg are, Calculate the image darkening threshold t_neg_1=I*λ_1, 0<λ_1<1; Calculate the dark area suppression coefficient λ_neg, λ_neg = y (i,j) 2 / (255+y (i,j) 2 ), where y (i,j) is the gray value corresponding to the position of the i-th row and j-th column of the image; Calculate the dimming threshold t_neg_2 = -1*t_neg_1*λ_neg; Δ_neg1=max(Δ_neg,t_neg_2).

5. The local contrast enhancement method according to claim 3 or 4, characterized in that: The specific steps for optimizing and brightening the image Δ_pos are: Calculate the image brightening threshold t_pos_1=(255-I)*λ_2, 0<λ_2<1; Calculate the bright area suppression coefficient λ_pos, λ_pos = (255-y (i,j) ) 2 / (255+(255-y (i,j) ) 2 ); Calculate the brightening threshold t_pos_2 = t_pos_1*λ_pos; Δ_pos1=max(Δ_pos,t_pos_2).

Images