Low-brightness image enhancement algorithm based on atmospheric scattering model

By calculating the maximum saturation enhancement value and the atmospheric light model, and combining it with the Retinex model, the brightness and saturation of low-brightness images are enhanced simultaneously. This solves the problem of poor visual effects of existing algorithms in atmospheric light scenes and enables wider application.

CN119048359AActive Publication Date: 2024-11-29NANJING UNIV OF SCI & TECH
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411103551.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-11-29
Estimated Expiration
2044-08-13

AI Technical Summary

Technical Problem

Existing low-brightness image enhancement algorithms ignore the role of saturation when improving brightness and removing noise, resulting in a hazy visual effect in low-brightness scenes with atmospheric light. Furthermore, deep learning-based algorithms have poor adaptability to scene changes, high hardware requirements, and limit their application scope.

Method used

By calculating the maximum saturation enhancement value of different color pixels in the original low-brightness image, a mathematical model of atmospheric light and saturation enhancement value is established. The atmospheric light is smoothed using norms, and brightness and saturation are simultaneously enhanced by combining the Retinex model.

Benefits of technology

It achieves saturation and brightness enhancement of low-brightness images under different environments, effectively removes hazy visual effects, and broadens the application scenarios of low-light image algorithms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048359A_ABST
    Figure CN119048359A_ABST
Patent Text Reader

Abstract

The invention discloses a low-brightness image enhancement algorithm based on an atmospheric scattering model. The low-brightness image enhancement algorithm comprises the following steps: calculating maximum saturation improvement values of different color pixels in an original low-brightness image; establishing a mathematical model between the atmospheric light and the saturation lifting value to estimate the atmospheric light of the original low-brightness image, and smoothing the estimated atmospheric light to obtain locally smooth atmospheric light; removing atmospheric light in the original low-brightness image to obtain an atmospheric light removed image; estimating a brightness map of the image by using the maximum brightness prior, and smoothing the estimated brightness map to obtain a smooth brightness map; and inputting the locally smooth atmospheric light and the smooth brightness graph into a retinex model to obtain a reflection image with synchronously enhanced brightness and saturation. According to the method, the saturation and the brightness of the low-brightness images in different environments can be enhanced, the visual effect of removing fog in a low-brightness scene containing atmospheric light of the image is effectively ensured, and the application scene of a low-illumination image algorithm is remarkably widened.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of visible light imaging and relates to a low-brightness image enhancement algorithm based on an atmospheric scattering model. Background Art

[0002] Due to the camera's own performance or the low-light environment, the captured image presents a low illumination. Low light blurs the details of the scene, destroys the original characteristics of the color, reduces the contrast of the image, and cannot provide a good visual experience. In addition, it will also affect the use of other image tasks, such as infrared and visible light image fusion, target detection and tracking, image segmentation, etc.

[0003] To solve this problem, scholars at home and abroad have conducted extensive research on low-brightness image enhancement algorithms and proposed a large number of effective algorithms. However, the existing low-brightness image algorithms still have some limitations:

[0004] (1) Traditional low-brightness image enhancement algorithms often focus on brightness enhancement, noise removal, and contrast enhancement, but ignore the role of saturation enhancement. When applied to low-brightness scenes with a small amount of atmospheric light, the enhancement results in a hazy visual feature due to the lack of saturation enhancement capability, which is difficult to meet the requirements of practical applications.

[0005] (2) The low-brightness enhancement algorithm based on deep learning has good brightness enhancement capabilities, but the saturation enhancement capability is limited. The effect of the algorithm based on supervised learning depends on paired input data. When the usage scenario changes, the adaptability of the algorithm will degrade. The algorithm based on unsupervised learning gets rid of the algorithm's reliance on paired data, but its effect is often not as good as that of the algorithm based on supervised learning. In addition, the low-brightness image enhancement algorithm based on deep learning has too high requirements on hardware, which limits the actual scope of application. Summary of the invention

[0006] Purpose of the invention: The purpose of the present invention is to provide a low-brightness image enhancement algorithm based on an atmospheric scattering model, which can enhance the saturation and brightness of low-brightness images in different environments.

[0007] Technical solution: To achieve the above purpose, the low-brightness image enhancement algorithm based on the atmospheric scattering model described in the present invention comprises the following steps:

[0008] Step 1: Calculate the maximum saturation enhancement value of different color pixels in the original low-brightness image;

[0009] Step 2: Establish a mathematical model between atmospheric light and saturation enhancement value to estimate the atmospheric light Q of the original low-brightness image, and use the norm to smooth the estimated atmospheric light Q to obtain a locally smooth atmospheric light Q. o ;

[0010] Step 3: Using the locally smooth atmospheric light Q o , remove the atmospheric light from the original low-brightness image, and obtain the image S1 without atmospheric light; use the maximum brightness prior to estimate the brightness map t of image S1, and use the norm to smooth the estimated brightness map t to obtain a smooth brightness map t o ;

[0011] Step 4: Smooth the local atmospheric light Q o and a smooth brightness map t o Input into the retinex model to obtain a reflection image with synchronously enhanced brightness and saturation.

[0012] Preferably, the original low-brightness image in step 1 is a visible light image having RGB channels.

[0013] Preferably, the maximum saturation enhancement value of different color pixels in the original low-brightness image described in step 1 is calculated using an image color classification algorithm, including establishing a mathematical model of the maximum saturation enhancement value, and using the mathematical model to calculate the color category w i Maximum saturation boost value

[0014] Preferably, the mathematical model of the maximum saturation improvement value is shown as follows:

[0015]

[0016] In the formula, S represents the original low-brightness image, k means represents the k-means classification algorithm, w i Indicates color category. The original low-brightness image has n types of colors, represented by w1, w2, … w n express; Indicates that the color category is w i The saturation of the pixel.

[0017] Preferably, the mathematical model between the atmospheric light and the saturation enhancement value is established in step 2 to estimate the atmospheric light Q of the original low-brightness image, specifically:

[0018] The mathematical model between atmospheric light and maximum saturation boost value is as follows:

[0019]

[0020] In the formula, S max Indicates the maximum value of the pixel in the RGB channel of the original low-brightness image, ΔS satIt represents the maximum saturation improvement value of all color categories in the original low-brightness image, k0 represents the atmospheric light adjustment factor, M represents the ratio of the minimum pixel value to the maximum pixel value of the RGB channel of the original low-brightness image, k is an adjustment factor between 0 and 1, and ε is a constant to prevent the denominator from being 0.

[0021] Preferably, the norm model in step 2 smoothes the estimated atmospheric light as shown in the following formula:

[0022]

[0023] In the formula, λ1 represents the constraint factor, and its value range is 0-1.

[0024] Preferably, the image S1 in step 3 where the atmospheric light is removed is SQ o , S represents the original low-brightness image.

[0025] Preferably, the brightness map of the image S1 in step 3 is t=max rgb (S1).

[0026] Preferably, the norm model in step 3 smoothes the estimated brightness map t as shown in the following formula:

[0027]

[0028] In the formula, λ1 represents the constraint factor, which ranges from 0 to 1.

[0029] Preferably, the step 4 described in which the locally smooth atmospheric light Q o and a smooth brightness map t o Input into the retinex model, and the reflection image with synchronously enhanced brightness and saturation is obtained as shown in the following formula:

[0030] Q o =R*t o

[0031] Where R represents the output image with high saturation and brightness after the original low-brightness image is enhanced.

[0032] Beneficial effect: The algorithm described in the present invention can enhance the saturation and brightness of low-brightness images in different environments, effectively ensure that the image removes the hazy visual effect in low-brightness scenes containing atmospheric light, and significantly broaden the application scenarios of low-light image algorithms. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a flow chart of the low brightness image enhancement algorithm based on the atmospheric scattering model of the present invention;

[0034] Figure 2The figure is a specific algorithm flow chart of the low brightness image enhancement algorithm based on the atmospheric scattering model of the present invention. DETAILED DESCRIPTION

[0035] The present invention will now be described in further detail with reference to the accompanying drawings.

[0036] See also Figure 1 and Figure 2 As shown, the low brightness image enhancement algorithm based on the atmospheric scattering model of the present invention comprises the following steps:

[0037] S1. Calculate the maximum saturation enhancement value of different color pixels in the original low-brightness image using an image color classification algorithm. The low-brightness image is a visible light image with RGB channels, and the atmospheric light in the environment of the low-brightness image is very weak. Specifically:

[0038] A mathematical model of the maximum saturation boost value is established, which is expressed as follows:

[0039]

[0040] Where S refers to the original low-brightness image, k means Refers to the k-means classification algorithm, w i Refers to color category. The original low-brightness image is divided into n color categories, respectively w1, w2, …w n express. The color category is w i The saturation of the pixel, Refers to the color category as w i Maximum saturation.

[0041] S2. Establish a mathematical model between atmospheric light and saturation enhancement value to estimate the atmospheric light Q of the original low-brightness image, and use the norm to smooth the estimated atmospheric light Q to obtain a locally smooth atmospheric light Q. o , specifically:

[0042] S21. Establish a mathematical model between atmospheric light and maximum saturation enhancement value, as shown in the following formula:

[0043]

[0044] In the formula, S max Indicates the maximum value of the pixel in the RGB channel of the original low-brightness image, ΔS satIt represents the maximum saturation enhancement value of all color categories in the original low-brightness image, k0 represents the atmospheric light adjustment factor, M represents the ratio of the minimum pixel value to the maximum pixel value of the RGB channel of the original low-brightness image, k is an adjustment factor between 0 and 1, and ε is a very small constant to prevent the denominator from being 0, which is set to 0.00001 here.

[0045] S21. Use the norm model to smooth the estimated atmospheric light and obtain the locally smooth atmospheric light, as shown in the following formula:

[0046]

[0047] Where λ1 is the constraint factor, and its range is 0-1.

[0048] S3, using locally smooth atmospheric light Q o , remove the atmospheric light from the original low-brightness image, and obtain the image S1 without atmospheric light; use the maximum brightness prior to estimate the brightness map t of image S1, and use the norm to smooth the estimated brightness map t to obtain a smooth brightness map t o

[0049] S31, image S1 with atmospheric light removed:

[0050] S1=SQ o

[0051] S32, based on S1, the brightness map is extracted using the maximum brightness prior, as shown in the following formula:

[0052] t=max rgb (S1)

[0053] S33, smoothing the estimated brightness map using the norm model, as shown in the following formula:

[0054]

[0055] Where, t o Refers to the smoothed brightness map of the output.

[0056] S4, the local smooth atmospheric light Q o and a smooth brightness map t o Input into the retinex model to obtain a reflection image with synchronously enhanced brightness and saturation, specifically:

[0057] Q o =R*t o

[0058] Where R is the output of the enhanced original low-brightness image, which is an output image with high saturation and brightness.

Claims

1. A low brightness image enhancement algorithm based on an atmospheric scattering model, characterized in that: The following steps are involved: Step 1: Calculate the maximum saturation enhancement value of different color pixels in the original low-brightness image; Step 2: Establish a mathematical model between atmospheric light and saturation enhancement value to estimate the atmospheric light Q of the original low-brightness image, and use the norm to smooth the estimated atmospheric light Q to obtain a locally smooth atmospheric light Q. o ; Step 3: Using the locally smooth atmospheric light Q o , remove the atmospheric light from the original low-brightness image, and obtain the image S1 without atmospheric light; use the maximum brightness prior to estimate the brightness map t of image S1, and use the norm to smooth the estimated brightness map t to obtain a smooth brightness map t o ; Step 4: Q o and a smooth brightness map t o Input into the retinex model to obtain a reflection image with synchronously enhanced brightness and saturation.

2. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: The original low-brightness image described in step 1 is a visible light image having RGB channels.

3. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: The maximum saturation improvement value of different color pixels in the original low-brightness image described in step 1 is calculated using an image color classification algorithm, including establishing a mathematical model of the maximum saturation improvement value, and using the mathematical model to calculate the color category w i Maximum saturation boost value 4. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 3 is characterized in that: The mathematical model of the maximum saturation improvement value is shown in the following formula: Where S represents the original low-brightness image, k means represents the k-means classification algorithm, w i Indicates color category. The original low-brightness image has n types of colors, represented by w1, w2, ...w n express; Indicates that the color category is w i The saturation of the pixel.

5. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: In step 2, a mathematical model between atmospheric light and saturation enhancement value is established to estimate the atmospheric light Q of the original low-brightness image, specifically: The mathematical model between atmospheric light and maximum saturation boost value is as follows: In the formula, S max Indicates the maximum value of the pixel in the RGB channel of the original low-brightness image, ΔS sat It represents the maximum saturation improvement value of all color categories in the original low-brightness image, k0 represents the atmospheric light adjustment factor, M represents the ratio of the minimum pixel value to the maximum pixel value of the RGB channel of the original low-brightness image, k is an adjustment factor between 0 and 1, and ε is a constant to prevent the denominator from being 0.

6. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: The norm model in step 2 smoothes the estimated atmospheric light as shown in the following formula: In the formula, λ1 represents the constraint factor, and its value range is 0-1.

7. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: The image S1 = SQ in step 3 where atmospheric light is removed o , S represents the original low-brightness image.

8. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: The brightness map of the image S1 in step 3 is t=max rgb (S1).

9. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: The norm model in step 3 smoothes the estimated brightness map t as shown in the following formula: In the formula, λ1 represents the constraint factor, which ranges from 0 to 1.

10. The low brightness image enhancement algorithm based on the atmospheric scattering model according to claim 1, characterized in that: The local smooth atmospheric light Q o and a smooth brightness map t o Input into the retinex model, and the reflection image with synchronously enhanced brightness and saturation is obtained as shown in the following formula: Q o =R*t o Where R represents the output image with high saturation and brightness after the original low-brightness image is enhanced.

Citation Information

Patent Citations

  • Rapid image defogging method based on brightness contrast enhancement and saturation compensation

    CN108416741A

  • Low-illumination color image real-time enhancement method and system

    CN112488957A

  • Method for determining if an input image is a foggy image, method for determining a foggy level of an input image and cleaning method for foggy images

    US20110135200A1