Airborne imaging non-uniform low-light scene image enhancement method

Abstract

The present application belongs to the technical field of image processing, and particularly relates to an airborne imaging non-uniform low-light scene image enhancement method. The method comprises the following steps: S1, processing a non-uniform low-light scene image to obtain H, S and V channel images; S2, performing dark area correction calculation on the V channel image to obtain a first image; S3, performing bright area suppression calculation on the V channel image to obtain a second image; S4, fusing the first and second images to obtain a fused image; S5, inputting a pixel distribution optimization image into a nonlinear image transformation tone adjustment model for processing to obtain a mid-term enhancement image; and S6, performing pixel value range stretching processing on the mid-term enhancement image, and combining the H and S channel images to obtain an optimized enhancement image. The present application can effectively improve the visibility of the dark area of the image and retain the detail feature information of the bright area while retaining the color richness and level of the image through the regional differentiation calculation and adaptive fusion mechanism.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology, and in particular relates to an image enhancement method for non-uniform low-light scenes in airborne imaging. Background Technology

[0002] High-quality images are the foundation and key guarantee for the implementation of airborne imaging applications. When conducting airborne imaging acquisition in low-light and unevenly lit scenarios, the inherent limitations of the dynamic range of airborne imaging equipment mean that conventional methods, such as adjusting the exposure time, cannot simultaneously solve the problems of underexposure in dark areas and overexposure in bright areas. This makes it difficult to directly obtain high-quality images with clear details in both bright and dark areas and good color visual effects. Therefore, adaptive enhancement research on images acquired in non-uniform low-light scenarios for airborne imaging is an important technical issue that urgently needs to be addressed. This research can not only effectively support airborne remote sensing imaging but also support technological upgrades and development in multiple fields such as medical imaging, intelligent transportation, and security monitoring.

[0003] To improve the visual quality of images in non-uniform low-light scenes in airborne imaging, researchers have proposed various enhancement methods, mainly as follows: Brightness stretching is achieved by adjusting the histogram distribution. While this method is computationally simple and easy to apply in engineering, it is sensitive to noise and prone to over-enhancement and color degradation. Retinex-based methods decompose the image into illumination and reflection components. Estimating the illumination component can improve contrast, but it easily leads to color distortion. Deep learning-based methods rely on convolutional neural networks for high-performance enhancement, but they depend on the quality of training data, have limited generalization ability, and are resource-intensive, making them difficult to deploy on resource-constrained devices. In summary, existing technologies generally suffer from insufficient universality and weak non-uniform illumination processing capabilities. Global image enhancement methods struggle to simultaneously enhance dark details and suppress highlights, resulting in enhanced images prone to local overexposure, underexposure, color distortion, and loss of detail. Furthermore, existing technologies have limited adaptive processing capabilities for scenes with extremely uneven illumination distribution, failing to effectively improve the visibility of dark areas while fully preserving the texture information of bright areas. Summary of the Invention

[0004] In view of this, the present invention aims to provide an image enhancement method for non-uniform low-light scenes in airborne imaging, in order to solve the problems of poor universality and insignificant enhancement effect on non-uniform low-light scene images in the existing technology. The present invention effectively improves the visibility of dark areas and retains the detailed feature information of bright areas in the image while preserving the richness of image color and the sense of layering through regional differential calculation and adaptive fusion mechanism.

[0005] To achieve the above objectives, the technical solution created by this invention is implemented as follows:

[0006] An image enhancement method for non-uniform low-light scenes in airborne imaging specifically includes the following steps:

[0007] S1: Process the non-uniform low-light scene image to obtain the H channel image, S channel image and V channel image of the non-uniform low-light scene image;

[0008] S2: Perform dark area correction calculations on the V channel image to obtain the first image;

[0009] S3: Perform brightness suppression calculation on the V channel image to obtain the second image;

[0010] S4: Merge the first image and the second image to obtain a merged image;

[0011] S5: Stretch the pixel value range of the fused image to obtain a pixel distribution optimized image, construct a nonlinear image transformation tone adjustment model based on statistical features, input the pixel distribution optimized image into the nonlinear image transformation tone adjustment model for processing, and obtain the mid-term enhanced image;

[0012] S6: Perform pixel value range stretching on the mid-term enhanced image and combine the H channel image and S channel image to obtain an optimized enhanced image.

[0013] Furthermore, the non-uniform low-light scene image is an RGB image.

[0014] Furthermore, step S1 specifically includes:

[0015] S11: Normalize the pixel values ​​of the pixels contained in the non-uniform low-light scene image, and calculate the global intensity mean of the normalized image.

[0016] S12: Convert the normalized image from the RGB color space to the HSV color space, and separate the converted image to obtain the H channel image, S channel image, and V channel image.

[0017] Furthermore, step S2 specifically includes the following steps:

[0018] S21: Perform exponential mapping pre-enhancement calculation on the V channel image to obtain a brightness-enhanced image;

[0019] S22: An improved inverse hyperbolic sine Naka-Rushton nonlinear mapping model is used to adaptively enhance the brightness-upgraded image to obtain the first image:

[0020] ;

[0021] Where U is the first image, A is the brightness-enhanced image, asinh() is the inverse hyperbolic sine function, and c is a constant.

[0022] Furthermore, in step S3, an improved exponentially modified cumulative distribution function is used to adaptively suppress bright areas in the V channel image, resulting in a second image:

[0023] ;

[0024] ;

[0025] Where O is the second image, b is the adaptive suppression parameter, and atan( () represents the arctangent function, V represents the V channel image, and e is the natural constant. This represents the global average intensity.

[0026] Furthermore, in step S4, the first image and the second image are weighted and fused based on a pixel-level adaptive remapping model to obtain a fused image:

[0027] ;

[0028] ;

[0029] Where F is the fused image, U is the first image, O is the second image, c is a constant, and f is the adaptive region fusion coefficient. This represents the global average intensity.

[0030] Furthermore, step S5 specifically includes the following steps:

[0031] S51: Obtain a pixel distribution optimized image by stretching the pixel value range of the fused image using the following formula:

[0032] ;

[0033] Where F represents the fused image, and P represents the pixel distribution optimized image. To merge the minimum pixel values ​​in the image, To merge the maximum pixel values ​​in the image;

[0034] S52: Construct a nonlinear image transformation tone adjustment model based on statistical features, and calculate the pixel distribution optimization image using the following formula to obtain the mid-term enhanced image:

[0035] ;

[0036] ;

[0037] Where Q represents the mid-term enhanced image, The standard deviation of the mid-term enhanced image, The variance of the image during mid-term enhancement is represented by α, which is an adaptive adjustment parameter. This represents the global average intensity.

[0038] Furthermore, step S6 specifically includes the following steps:

[0039] S61: The pixel value range stretching process of the mid-term enhanced image is performed using the following formula to obtain the brightness channel image:

[0040] ;

[0041] in, For the brightness channel image, The maximum pixel value for mid-term image enhancement. This represents the minimum pixel value for the mid-term enhanced image.

[0042] S62: Converts the luminance channel image, H channel image, and S channel image from the HSV color space to the RGB color space to obtain an optimized and enhanced image.

[0043] Compared with the prior art, the present invention can achieve the following beneficial effects:

[0044] This invention presents an image enhancement method for non-uniform low-light scenes in airborne imaging. By correcting and suppressing the dark and bright areas of the V channel in the non-uniform low-light scene image, and through adaptive bright-dark area fusion reconstruction, tone adjustment calculation, pixel value range stretching, and color space transformation, a bright and clear image is obtained. This invention effectively preserves the detailed features of both dark and bright areas while maintaining the natural and vibrant colors and sense of depth of the image. It effectively expands the dynamic range of the image while ensuring color fidelity and visual depth of the enhanced image. This invention has low computational complexity, robust and efficient performance, and strong generalization ability. It is applicable not only to airborne remote sensing imaging but also to various non-uniform low-light scenes such as mobile terminal photography and surveillance imaging, demonstrating good engineering adaptability. Attached Figure Description

[0045] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0046] Figure 1 This is a schematic flowchart of the airborne imaging non-uniform low-light scene image enhancement method according to an embodiment of the present invention;

[0047] Figure 2 The first actual airborne ground imaging image and the corresponding RGB three-channel color distribution curve of the first actual airborne ground imaging image are described in the embodiments of the present invention.

[0048] Figure 2 (a) A first actual airborne ground imaging image as described in an embodiment of the present invention;

[0049] Figure 2 (b) The RGB three-channel color distribution curve diagram corresponding to the first actual airborne ground imaging map as described in the embodiment of the present invention;

[0050] Figure 3 The first effect image and the corresponding RGB three-channel color distribution curve of the first effect image obtained by adaptive enhancement of the first actual airborne ground imaging image as described in the embodiment of the present invention;

[0051] Figure 3 (a) A first effect image obtained by adaptively enhancing a first actual airborne ground imaging image as described in an embodiment of the present invention;

[0052] Figure 3 (b) The RGB three-channel color distribution curve diagram corresponding to the first effect diagram as described in the embodiment of the present invention;

[0053] Figure 4 The RGB three-channel color distribution curve diagram corresponding to the second actual airborne ground imaging image and the second actual aerial ground imaging image described in the embodiments of the present invention;

[0054] Figure 4 (a) A second actual airborne ground imaging image as described in the embodiment of the present invention;

[0055] Figure 4 (b) The RGB three-channel color distribution curve diagram corresponding to the second actual airborne ground imaging map as described in the embodiment of the present invention;

[0056] Figure 5 The second effect image obtained by adaptively enhancing the second actual airborne ground imaging image as described in the embodiment of the present invention, and the color distribution curve of the RGB three channels corresponding to the second effect image;

[0057] Figure 5 (a) A second effect image obtained by adaptively enhancing a second actual airborne ground imaging image as described in an embodiment of the present invention;

[0058] Figure 5 (b) The RGB three-channel color distribution curve diagram corresponding to the second effect diagram as described in the embodiment of the present invention;

[0059] Figure 6 The everyday scene imaging image and the corresponding RGB three-channel color distribution curve of the everyday scene imaging image are described in the embodiments of the present invention;

[0060] Figure 6 (a) An image of a daily life scene as described in an embodiment of the present invention;

[0061] Figure 6 (b) An RGB three-channel color distribution curve diagram corresponding to the image of a daily life scene as described in the embodiment of the present invention;

[0062] Figure 7 The third effect image and the corresponding RGB three-channel color distribution curve of the third effect image obtained by adaptive enhancement of the image of daily life scene as described in the embodiment of the present invention;

[0063] Figure 7 (a) A third effect image obtained by adaptively enhancing an image of a daily life scene as described in an embodiment of the present invention;

[0064] Figure 7 (b) The RGB three-channel color distribution curve diagram corresponding to the third effect diagram as described in the embodiment of the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0066] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0067] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0068] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0069] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0070] like Figure 1 As shown, this invention proposes an image enhancement method for airborne imaging in non-uniform low-light scenes, specifically including the following steps:

[0071] S1: Process the non-uniform low-light scene image to obtain the H channel image, S channel image and V channel image of the non-uniform low-light scene image;

[0072] S2: Perform dark area correction calculations on the V channel image to obtain the first image;

[0073] S3: Perform brightness suppression calculation on the V channel image to obtain the second image;

[0074] S4: Merge the first image and the second image to obtain a merged image;

[0075] S5: Stretch the pixel value range of the fused image to obtain a pixel distribution optimized image, construct a nonlinear image transformation tone adjustment model based on statistical features, input the pixel distribution optimized image into the nonlinear image transformation tone adjustment model for processing, and obtain the mid-term enhanced image;

[0076] S6: Perform pixel value range stretching on the mid-term enhanced image and combine the H channel image and S channel image to obtain an optimized enhanced image.

[0077] It should be noted that: S1: Calculate the H-channel, S-channel, and V-channel images of the input non-uniform low-light scene image; S2: Perform dark area correction calculation on the V-channel image to enhance the details of the dark areas, obtaining the first image U; S3: Perform bright area suppression calculation on the V-channel image to preserve the details of the bright areas, obtaining the second image O; S4: Fuse the first image U and the second image O to obtain useful brightness complementary information, obtaining the fused image F; S5: Stretch the pixel value range of the fused image F, establish a nonlinear image transformation tone adjustment model based on statistical features, and obtain an intermediate enhancement image Q with clear dark and bright areas; S6: Stretch the pixel value range of the intermediate enhancement image Q, and perform color space transformation by combining the original S-channel and H-channel images to obtain the final optimized enhancement image.

[0078] This invention effectively solves the problem of simultaneously improving the brightness of dark areas and maintaining the detail of bright areas in images of non-uniform low-light scenes, while preserving color fidelity and sense of layering, without color distortion or degradation. It is also simple to calculate and easy to deploy.

[0079] In some embodiments, the airborne imaging non-uniform low-light scene image is an RGB image.

[0080] In some embodiments, step S1 specifically includes:

[0081] S11: Normalize the pixel values ​​of the pixels contained in the non-uniform low-light scene image, and calculate the global intensity mean of the normalized image.

[0082] S12: Convert the normalized image from the RGB color space to the HSV color space, and separate the converted image to obtain the H channel image, S channel image, and V channel image.

[0083] It should be noted that the input is a non-uniform low-light scene image, the pixel values ​​of each pixel in the image are normalized to obtain a normalized color image, and the global average brightness of the normalized image is calculated:

[0084]

[0085] Among them, I m H is the global intensity mean, H is the height of the normalized image, W is the width of the normalized image, i is the row count, j is the column count, and k is the color channel count.

[0086] The normalized image is converted from the RGB color space to the HSV color space, and the hue (H channel), saturation (S channel), and lightness (V channel) of the image are separated.

[0087] In some embodiments, step S2 specifically includes the following steps:

[0088] S21: Perform exponential mapping pre-enhancement calculation on the V channel image to obtain a brightness-enhanced image;

[0089] S22: An improved inverse hyperbolic sine Naka-Rushton nonlinear mapping model is used to adaptively enhance the brightness-upgraded image to obtain the first image:

[0090] ;

[0091] Where U is the first image, A is the brightness-enhanced image, asinh() is the inverse hyperbolic sine function, and c is a constant.

[0092] It should be noted that, keeping the components of the H-channel and S-channel images unchanged, an exponential mapping pre-enhancement calculation is performed on the input brightness V-channel image to obtain the preliminary brightness-enhanced image A:

[0093] ;

[0094] in, As a brightness enhancement parameter, considering both the brightness and contrast quality of the luminance V channel image of a non-uniform low-light scene, the value of τ is set to 0.9.

[0095] An improved inverse hyperbolic sine Naka-Rushton nonlinear mapping model is used to adaptively enhance the brightness of the enhanced image A, resulting in a first image U with improved detail brightness in the dark areas:

[0096] ;

[0097] Wherein, asinh() is the inverse hyperbolic sine function, log2() is the logarithm calculation with base 2, and c is the minimum constant to prevent division by zero, used to avoid data anomalies.

[0098] In some embodiments, in step S3, an improved exponentially modified cumulative distribution function is used to adaptively suppress bright areas in the V channel image to obtain a second image:

[0099] ;

[0100] ;

[0101] Where O is the second image, b is the adaptive suppression parameter, and atan( () represents the arctangent function, V represents the V channel image, and e is the natural constant. This represents the global average intensity.

[0102] It should be noted that for the bright areas of the brightness V channel image, an improved exponentially modified cumulative distribution function is used to perform nonlinear adaptive correction on the V channel image. This step can compress the dynamic range of the brightness V channel image, making the dark areas of the brightness V channel image darker and reducing the amplitude of the bright areas of the brightness V channel image, thus avoiding overexposure and achieving image bright area suppression.

[0103] ;

[0104] ;

[0105] Where O represents the second image; b is the adaptive suppression parameter, and the larger the b, the stronger the protection of bright areas; atan( ) is the arctangent function; to better adapt to various scenarios, the present invention designs the value of b to be adaptively adjustable, specifically calculated based on the global average brightness of the input image.

[0106] In some embodiments, in step S4, the first image and the second image are weighted and fused based on a pixel-level adaptive remapping model to obtain a fused image:

[0107] ;

[0108] ;

[0109] Where F is the fused image, U is the first image, O is the second image, c is a constant, and f is the adaptive region fusion coefficient. This represents the global average intensity.

[0110] It should be noted that, after the preceding calculations, a first image U with enhanced brightness in the aforementioned dark areas and a second image O with suppressed brightness in the aforementioned bright areas can be obtained. Based on a pixel-level adaptive remapping model, the first image U and the second image O are weighted and fused to obtain a fused image F with complementary brightness information.

[0111] ;

[0112] ;

[0113] Where f is the adaptive region fusion coefficient, which adapts to local brightness characteristics and ensures a natural brightness transition in the middle region. The f value designed in this invention is adaptive and adjustable, and is calculated based on the global average brightness of the input image.

[0114] In some embodiments, step S5 specifically includes the following steps:

[0115] S51: Obtain a pixel distribution optimized image by stretching the pixel value range of the fused image using the following formula:

[0116] ;

[0117] Where F represents the fused image, and P represents the pixel distribution optimized image. To merge the minimum pixel values ​​in the image, To merge the maximum pixel values ​​in the image;

[0118] S52: Construct a nonlinear image transformation tone adjustment model based on statistical features, and calculate the pixel distribution optimization image using the following formula to obtain the mid-term enhanced image:

[0119] ;

[0120] ;

[0121] Where Q represents the mid-term enhanced image, The standard deviation of the mid-term enhanced image, The variance of the image during mid-term enhancement is represented by α, which is an adaptive adjustment parameter. This represents the global average intensity.

[0122] It should be noted that, based on the maximum and minimum pixel values ​​of the fused image, the pixel value range of the fused image F is stretched to obtain the pixel distribution optimized image P;

[0123] ;

[0124] Among them, F n To merge the minimum pixel value in image F, F m This is to merge the maximum pixel values ​​in image F.

[0125] A nonlinear image transformation tone adjustment model based on statistical features is established to obtain a mid-term enhanced image Q with clear details in both dark and bright areas.

[0126] ;

[0127] ;

[0128] Where, σ P and u P Let α be the standard deviation and variance of the pixel distribution optimization image P, and let α be the adjustment parameter. In this invention, the value of α is designed to be adaptive and adjustable, and is calculated based on the global average brightness of the input image.

[0129] In some embodiments, step S6 specifically includes the following steps:

[0130] S61: The pixel value range stretching process of the mid-term enhanced image is performed using the following formula to obtain the brightness channel image:

[0131] ;

[0132] in, For the brightness channel image, The maximum pixel value for mid-term image enhancement. This represents the minimum pixel value for the mid-term enhanced image.

[0133] S62: Converts the luminance channel image, H channel image, and S channel image from the HSV color space to the RGB color space to obtain an optimized and enhanced image.

[0134] It should be noted that, based on the minimum and maximum pixel values ​​of the Q-value of the mid-term enhanced image, the dynamic range of pixel values ​​is optimized to better align with human visual habits.

[0135] ;

[0136] Among them, Q n Q is the minimum pixel value in the intermediate-stage enhanced image Q. m To obtain the maximum pixel value in the intermediate enhancement image Q, the above operation can rearrange the pixel values ​​in the intermediate enhancement image Q and map them to the range of [0,1] to obtain a refined brightness channel image Z.

[0137] Based on the original S-channel, H-channel, and luminance channel image Z, an enhanced HSV image is obtained. This image is then subjected to an inverse HSV-to-RGB color space transformation to obtain the final optimized and enhanced image with clear brightness and vivid colors.

[0138] The image enhancement method for non-uniform low-light scenes in airborne imaging provided by this invention has the following technical advantages compared with existing technologies: In terms of adaptive enhancement, based on the global intensity mean and local statistical features of the input image, the enhancement degree of dark areas and the suppression parameters of bright areas are adaptively adjusted to achieve dynamic optimization of the enhancement degree and improve the algorithm's adaptability to different lighting scenes. In terms of maintaining a wide dynamic range, through regional differential processing and an adaptive fusion mechanism, the visibility of details in dark areas is improved while effectively suppressing overexposure in bright areas, significantly expanding the image's dynamic range and improving the imaging effect under non-uniform lighting. In terms of color preservation, the brightness channel is processed independently in the HSV color space to maintain hue and saturation information, fundamentally avoiding problems such as color shift, saturation distortion, and color degradation. The image enhanced by this invention is bright, clear, and has natural and vibrant colors, effectively improving the visual quality and dynamic range of images in non-uniform low-light scenes in airborne imaging. The model has low computational cost and can provide technical support for obtaining high dynamic range imaging in low-light or non-uniform lighting scenes. Figure 2 (a) and Figure 3 (a) Figure 4(a) and Figure 5 (a) Figure 6 (a) and Figure 7 (a) It can be seen that by applying this invention, the brightness of dark areas in non-uniform low-light scene images can be significantly improved, while details in bright areas are preserved, and the images possess vibrant colors and rich gradations without color degradation or other problems, thus significantly improving the visual quality of the images. And from the comparison... Figure 2 (b) and Figure 3 (b) Figure 4 (b) and Figure 5 (b) Figure 6 (b) and Figure 7 (b) It can be seen that the peak values ​​of the R, G and B pixel values ​​of the enhanced image obtained by applying the present invention move from the left end of the horizontal axis of the pixel value to the right end of the horizontal axis, and the distribution range of the three colors is wider and the differentiation is better.

[0139] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0140] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An airborne imaging non-uniform low-light scene image enhancement method, characterized by: Specifically comprising the following steps: S1: processing the non-uniform low-light scene image to obtain an H channel image, an S channel image and a V channel image of the non-uniform low-light scene image; S2: performing dark area correction calculation on the V channel image to obtain a first image; S3: performing bright area suppression calculation on the V channel image to obtain a second image; S4: fusing the first image and the second image to obtain a fused image; S5: stretching the pixel value range of the fused image to obtain a pixel distribution optimized image, constructing a non-linear image transformation tone adjustment model based on statistical characteristics, inputting the pixel distribution optimized image into the non-linear image transformation tone adjustment model for processing to obtain a mid-term enhanced image; S6: performing pixel value range stretching processing on the mid-term enhanced image, and combining the H channel image and the S channel image to obtain an optimized enhanced image.

2. The airborne imaging non-uniform low-light scene image enhancement method according to claim 1, characterized in that: The non-uniform low-light scene image is an RGB image.

3. The airborne imaging non-uniform low-light scene image enhancement method of claim 1, wherein: Step S1 specifically comprises: S11: performing normalization processing on the pixel values of the pixel points contained in the non-uniform low-light scene image, and calculating the global intensity mean value of the normalized image; S12: converting the normalized image from the RGB color space to the HSV color space, and separating the converted image to obtain the H channel image, the S channel image and the V channel image.

4. The airborne imaging non-uniform low-light scene image enhancement method of claim 1, wherein: Step S2 specifically comprises the following steps: S21: performing exponential mapping pre-enhancement calculation on the V channel image to obtain a luminance enhanced image; S22: performing adaptive enhancement on the luminance enhanced image using an improved inverse hyperbolic sine Naka-Rushton non-linear mapping model to obtain a first image: ； Wherein, U is the first image, A is the luminance enhanced image, asinh() is the inverse hyperbolic sine function, and c is a constant.

5. The airborne imaging non-uniform low-light scene image enhancement method of claim 1, wherein: In step S3, the V channel image is subjected to bright area adaptive suppression using an improved exponential correction cumulative distribution function to obtain a second image: ； ； where O is the second image, b is an adaptive suppression parameter, atan( ) is an arctangent function, V is the V channel image, e is the natural constant, is the global intensity mean.

6. The airborne imaging non-uniform low-light scene image enhancement method of claim 1, wherein: In step S4, the first image and the second image are weighted fused based on a pixel-level adaptive remapping model to obtain a fused image: ； ； Wherein, F is the fusion image, U is the first image, O is the second image, c is a constant, f is an adaptive region fusion coefficient, is the global intensity mean.

7. The airborne imaging non-uniform low-light scene image enhancement method of claim 1, wherein: Step S5 specifically comprises the following steps: S51: stretching the pixel value range of the fused image by the following formula to obtain a pixel distribution optimized image: ； wherein F is a fused image, P is a pixel distribution optimized image, is a minimum value of pixels in the fused image, is a maximum value of pixels in the fused image; S52: constructing a non-linear image transformation tone adjustment model based on statistical characteristics, calculating the pixel distribution optimized image by the following formula to obtain a mid-term enhanced image: ； ； wherein Q is the intermediate enhanced image, is the standard deviation of the intermediate enhanced image, is the variance of the intermediate enhanced image, and a is an adaptive adjustment parameter, is the global intensity mean.

8. The airborne imaging non-uniform low-light scene image enhancement method of claim 1, wherein: Step S6 specifically comprises the following steps: S61: performing pixel value range stretching processing on the mid-term enhanced image by the following formula to obtain a lightness channel image: ； wherein, is a luminance channel image, is a pixel maximum value of the intermediate enhanced image, is a pixel minimum value of the intermediate enhanced image; S62: converting the lightness channel image, the H channel image and the S channel image from the HSV color space to the RGB color space to obtain an optimized enhanced image.

Images