CN111861899A - A method and system for image enhancement based on uneven illumination - Google Patents

Abstract

The invention discloses an image enhancement method and system based on uneven illumination, which convert an original color image from an RGB space to an HSV space to obtain a brightness component V of the image; processing the illumination component V by using a path-optimized MR algorithm; and converting the processed HSV space into an RGB space, and synthesizing a new image. The advantages are that: the method is based on the image enhancement algorithm with uneven illumination, and the brightness component V of the image is processed by using the MR algorithm with optimized path, so that the brightness distribution of the image is more uniform. Compared with other algorithms, the algorithm has the advantages that the used paths are more uniform, and the brightness of the processed image is more reasonable in consideration of various different directions such as clockwise direction, anticlockwise direction and the like.

Description

A method and system for image enhancement based on uneven illumination

technical field

The invention relates to an image enhancement method and system based on uneven illumination, and belongs to the technical field of image processing.

Background technique

When an imaging device takes a photo, the quality of the photo is greatly affected by the light. The photos taken during the day are clearly visible, but the photos taken at night are more blurred. In addition, when acquiring images in underwater imaging, especially in deep water areas, artificial light sources are often required as auxiliary light sources for imaging, which results in uneven illumination and blurred details of the acquired images.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide an image enhancement method and system based on uneven illumination.

In order to solve the above technical problems, the present invention provides an image enhancement method based on uneven illumination, which converts the original color image from RGB space to HSV space, and obtains the brightness component V of the image; uses the path-optimized MR algorithm to process the brightness component V. ; Convert the processed HSV space to RGB space to synthesize a new image.

Further, the formula for converting the original color image from RGB space to HSV space is as follows:

in,

H represents the hue, V represents the brightness, S represents the saturation, R represents the pixel value of the red channel, G represents the pixel value of the green channel, B represents the pixel value of the blue channel, θ represents the rotation angle, and max() represents the maximum value , min() means take the minimum value.

Further, the MR algorithm processing process of the path optimization includes:

According to the global light and shade changes, determine several paths that describe the light and shade changes of the image with different starting points or different directions;

Select pixels on one of the paths, and use the selected pixels as part of the estimated luminance image;

The pixel value of the luminance component V is updated by utilizing the obtained central pixel point of the luminance component V and the light-dark contrast of a part of the pixel points of the estimated luminance image;

Iterate multiple times until the updated pixel value of the brightness component V contains the light and dark changes of the entire image, and is determined as the iterative pixel value of the brightness component V;

The iterative pixel value of the luminance component V of each path is determined, and the iterative pixel value of the luminance component V of all paths is summed and averaged to obtain the final pixel value of the luminance component V.

Further, the starting point of the path is (-offset, offset), the path is a spiral path covering the entire image, and the offset calculation formula is as follows:

为向下取整函数；rows和cols分别表示图像的行数和列数；Among them, offset represents the coordinates of the starting point,

is the round-down function; rows and cols represent the number of rows and columns of the image, respectively;

The update formula of the pixel value on the path is as follows:

Among them, n represents the number of iterations, r _n (x, y) represents the reflection information obtained by the nth iteration, I _n (x, y) represents the nth estimated brightness value,

Among them, max is the maximum value of the pixel in the original image, Δl=S _c -S _m is the brightness difference on the path, S _c represents the pixel value of the center point, S _m represents the pixel value of the point on the path, m=1,2,3 ....k represents a total of k pixels on the path.

Further, the formula for converting the processed HSV space to RGB space is as follows:

h _i =[H/60]mod6,f=H/60-h _i , p=V×(1-S), q=V×(1-f×S), t=V×(1-(1 -f)×S)

Among them, h _i , p, q, f, t are intermediate variables, and mod means remainder.

An image enhancement system based on uneven illumination, including:

The acquisition module is used to convert the original color image from RGB space to HSV space, and obtain the luminance component V of the image;

a processing module, used for processing the luminance component V using the MR algorithm of path optimization;

The synthesis module is used to convert the processed HSV space to RGB space and synthesize a new image.

Further, the acquisition module includes:

The first conversion module is used to convert the original color image from RGB space to HSV space by the following formula,

in,

H represents the hue, V represents the brightness, S represents the saturation, R represents the pixel value of the red channel, G represents the pixel value of the green channel, B represents the pixel value of the blue channel, θ represents the rotation angle, and max() represents the maximum value , min() means take the minimum value.

Further, the processing module includes:

The first determination module is used for determining several paths of the image light and dark changes with different starting points or different directions according to the global light and dark changes;

The selection module is used to select pixels on one of the paths, and use the selected pixels as a part of the estimated illuminance image;

The updating module is used to update the pixel value of the luminance component V by utilizing the obtained central pixel point of the luminance component V and the light-dark contrast of a part of the pixel point of the estimated illuminance image;

The second determination module is used for multiple iterations, until the updated pixel value of the brightness component V includes the light and dark changes of the entire picture, and is determined as the iterative pixel value of the brightness component V;

The third determination module is used for determining the iterative pixel value of the luminance component V of each path, and summing and averaging the iterated pixel values of the luminance component V of all paths to obtain the final pixel value of the luminance component V.

Further, the first determining module includes:

The first calculation module is used to calculate the starting point (-offset, offset) of the path,

The path is a spiral path covering the entire image, and the offset calculation formula is as follows:

为向下取整函数；rows和cols分别表示图像的行数和列数；Among them, offset represents the coordinates of the starting point,

is the round-down function; rows and cols represent the number of rows and columns of the image, respectively;

The update module includes:

The second calculation module is used to update the pixel value on the path by using the following formula,

Among them, n represents the number of iterations, r _n (x, y) represents the reflection information obtained by the nth iteration, I _n (x, y) represents the nth estimated brightness value,

Among them, max is the maximum value of the pixel in the original image, Δl=S _c -S _m is the brightness difference on the path, S _c represents the pixel value of the center point, S _m represents the pixel value of the point on the path, m=1,2,3 ....k represents a total of k pixels on the path.

Further, the synthesis module includes:

The second conversion module is used to convert the processed HSV space into RGB space by using the following formula,

h _i =[H/60]mod6,f=H/60-h _i , p=V×(1-S), q=V×(1-f×S), t=V×(1-(1 -f)×S)

Among them, h _i , p, q, f, t are intermediate variables, and mod means remainder.

Beneficial effects achieved by the present invention:

Based on the image enhancement algorithm of uneven illumination, the invention processes the brightness (V) of the image by using the MR algorithm of path optimization, so that the brightness distribution of the image is more uniform.

Compared with other algorithms, the path used by this algorithm is more uniform, and considering various different directions such as clockwise and counterclockwise, the brightness of the processed image is more reasonable.

Description of drawings

1 is a flowchart of an image enhancement algorithm based on uneven illumination of the present invention;

2 is a partial path diagram used by the MR algorithm in the present invention;

Figure 3 (a) is a clockwise path diagram used by the MR algorithm in the present invention, and Figure 3 (b) is a counterclockwise path diagram used by the MR algorithm in the present invention;

FIG. 4( a ) is the original image before image enhancement in the present invention, and FIG. 4( b ) is the image after image enhancement in the present invention.

Detailed ways

In order to make the purpose, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the following The described embodiments are only some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The invention discloses an image enhancement algorithm based on uneven illumination, which specifically includes the following steps

(1) Convert the original color image from RGB space to HSV space, and obtain the illuminance component V of the image;

(2) Use the path-optimized MR algorithm to process the luminance component V;

(3) Convert from HSV space to RGB space to synthesize new images.

The formula for converting the original color image from RGB space to HSV space in step (1) is as follows:

in,

The steps of the MR algorithm for path optimization in step (2) are as follows:

a Select a path that can describe the light and dark changes of the image according to the global light and shade changes;

b Select pixels on the specified path, and use the selected pixels as a part of the estimated luminance image;

c Use the contrast between the center pixel and the pixel on the path to update the pixel value;

d After many iterations, the center pixel value will contain the light and dark changes of the entire image.

In the present invention, the starting point of the path is (-offset, offset), where offset represents the distance from the target pixel, the initial distance is set as an index of 2, and the index part is smaller than the length and width of the input image, which is expressed as follows:

为向下取整函数；rows和cols分别表示图像的行数和列数。当确定了初始点(-offset,offset)后，随后需要确定的点为(offset,offset/2)、(offset/2,-offset)、 (-offset,-offset/2)。每次迭代时offset值较前一次减半，直到|offset|＜1迭代结束。最终得到图2的路径。In the above formula,

is the round-down function; rows and cols represent the number of rows and columns of the image, respectively. After the initial point (-offset,offset) is determined, the subsequent points to be determined are (offset,offset/2), (offset/2,-offset), (-offset,-offset/2). At each iteration, the offset value is halved from the previous one until |offset|<1 iteration ends. Finally, the path shown in Figure 2 is obtained.

When the corresponding point is selected, the points on the path need to be compared and updated. The formula is as follows:

In the above formula, n represents the number of iterations r _n (x, y), which represents the reflection information obtained by the nth iteration, I _n (x, y) represents the nth estimated brightness value, and n is 4 in the stupid experiment. In order to ensure that the maximum pixel value of the original image is not exceeded:

In the above formula, max is the maximum value of the pixel in the original image, Δl=S _c -S _m is the brightness difference on the path, S _c represents the pixel value of the center point, S _m represents the pixel value of the point on the path, m=1,2 ,3....k represents a total of k pixels on the path. Therefore, according to the constant comparison of the pixel difference on the path and the iterative operation, the unstable illuminance information value in the whole image is estimated through continuous iteration, so as to eliminate the influence of the illuminance information and obtain the reflection component value.

Although the path in Fig. 2 has covered the whole image, it does not take into account that the path of the farther pixels in the counterclockwise direction of the initial point includes clockwise and counterclockwise directions. The present invention performs the above paths in four directions respectively. The selection and completion of the iterative process is shown in Figures 3(a) and 3(b). Finally, the illuminance information obtained by the four different paths is averaged.

The formula for converting the color image from the HSV space to the RGB space in the step (3) is as follows:

h _i =[H/60]mod6,f=H/60-h _i , p=V×(1-S), q=V×(1-f×S), t=V×(1-(1 -f)×S)

In order to verify the effectiveness of this algorithm, multiple images are used to test the images before and after enhancement.

Figure 4(a) shows that the original image has the characteristics of blurring and uneven illumination. After the algorithm processing, as shown in Figure 4(b), the picture is clear and the image brightness is more uniform. Compared with the original image, the enhancement effect is remarkable.

It will be apparent to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments, but that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments are to be regarded in all respects as illustrative and not restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, which are therefore intended to fall within the scope of the appended claims. All changes within the meaning and scope of the same requirements are included in the present invention. Any reference signs in the claims shall not be construed as limiting the involved claim.

As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand: The technical solutions described in the embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions in the embodiments of the present invention.

Claims (10)

1. an image enhancement method based on uneven illumination, is characterized in that, Convert the original color image from RGB space to HSV space, and obtain the luminance component V of the image; Use the path-optimized MR algorithm to process the luminance component V; Convert the processed HSV space to RGB space to synthesize a new image. 2. The image enhancement method based on uneven illumination according to claim 1, wherein, The formula for converting the original color image from RGB space to HSV space is as follows:

in,

H represents the hue, V represents the brightness, S represents the saturation, R represents the pixel value of the red channel, G represents the pixel value of the green channel, B represents the pixel value of the blue channel, θ represents the rotation angle, and max() represents the maximum value , min() means take the minimum value. 3. The image enhancement method based on uneven illumination according to claim 1, wherein, The MR algorithm processing process of the path optimization includes: According to the global light and shade changes, determine several paths that describe the light and shade changes of the image with different starting points or different directions; Select pixels on one of the paths, and use the selected pixels as part of the estimated luminance image; The pixel value of the luminance component V is updated by utilizing the obtained central pixel point of the luminance component V and the light-dark contrast of a part of the pixel points of the estimated luminance image; Iterate multiple times until the updated pixel value of the brightness component V contains the light and dark changes of the entire image, and is determined as the iterative pixel value of the brightness component V; Determine the iterative pixel value of the luminance component V of each path, sum and average the iterated pixel values of the luminance component V of all paths, and obtain the final pixel value of the luminance component V. 4. The image enhancement method based on uneven illumination according to claim 3, wherein, The starting point of the path is (-offset, offset), and the path is a spiral path covering the entire image, where the offset calculation formula is as follows:

Among them, offset represents the coordinates of the starting point,

is the round-down function; rows and cols represent the number of rows and columns of the image, respectively; The update formula of the pixel value on the path is as follows:

Among them, n represents the number of iterations, r _n (x, y) represents the reflection information obtained by the nth iteration, I _n (x, y) represents the nth estimated brightness value,

Among them, max is the maximum value of the pixel in the original image, Δl=S _c -S _m is the brightness difference on the path, S _c represents the pixel value of the center point, S _m represents the pixel value of the point on the path, m=1, 2, 3 ....k represents a total of k pixels on the path. 5. The image enhancement method based on uneven illumination according to claim 1, wherein, The formula for converting the processed HSV space to RGB space is as follows:

h _i =[H/60]mod6,f=H/60-h _i , p=V×(1-S), q=V×(1-f×S), t=V×(1-(1 -f)×S) Among them, h _i , p, q, f, t are intermediate variables, and mod means remainder. 6. An image enhancement system based on uneven illumination, characterized in that, comprising: The acquisition module is used to convert the original color image from RGB space to HSV space, and obtain the luminance component V of the image; a processing module, used for processing the luminance component V using the MR algorithm of path optimization; The synthesis module is used to convert the processed HSV space to RGB space and synthesize a new image. 7. The image enhancement system based on uneven illumination according to claim 6, wherein the acquisition module comprises: The first conversion module is used to convert the original color image from RGB space to HSV space by the following formula,

in,

H represents the hue, V represents the brightness, S represents the saturation, R represents the pixel value of the red channel, G represents the pixel value of the green channel, B represents the pixel value of the blue channel, θ represents the rotation angle, and max() represents the maximum value , min() means take the minimum value. 8. The image enhancement system based on uneven illumination according to claim 6, wherein the processing module comprises: The first determination module is used for determining several paths of the image light and dark changes with different starting points or different directions according to the global light and dark changes; The selection module is used to select pixels on one of the paths, and use the selected pixels as a part of the estimated illuminance image; The updating module is used to update the pixel value of the luminance component V by utilizing the obtained central pixel point of the luminance component V and the light-dark contrast of a part of the pixel point of the estimated illuminance image; The second determination module is used for multiple iterations, until the updated pixel value of the brightness component V includes the light and dark changes of the entire picture, and is determined as the iterative pixel value of the brightness component V; The third determination module is used for determining the iterative pixel value of the luminance component V of each path, and summing and averaging the iterated pixel values of the luminance component V of all paths to obtain the final pixel value of the luminance component V. 9. The image enhancement system based on uneven illumination according to claim 8, wherein the first determining module comprises: The first calculation module is used to calculate the starting point (-offset, offset) of the path, The path is a spiral path covering the entire image, and the offset calculation formula is as follows:

Among them, offset represents the coordinates of the starting point,

is the round-down function; rows and cols represent the number of rows and columns of the image, respectively; The update module includes: The second calculation module is used to update the pixel value on the path by using the following formula,

Among them, n represents the number of iterations, r _n (x, y) represents the reflection information obtained by the nth iteration, I _n (x, y) represents the nth estimated brightness value,

Among them, max is the maximum value of the pixel in the original image, Δl=S _c -S _m is the brightness difference on the path, S _c represents the pixel value of the center point, S _m represents the pixel value of the point on the path, m=1,2,3 ....k represents a total of k pixels on the path. 10. The image enhancement system based on uneven illumination according to claim 1, wherein, The synthesis module includes: The second conversion module is used to convert the processed HSV space into RGB space by using the following formula,

h _i =[H/60]mod6,f=H/60-h _i , p=V×(1-S), q=V×(1-f×S), t=V×(1-(1 -f)×S) Among them, h _i , p, q, f, t are intermediate variables, and mod means remainder.

Images