An Adaptive Dynamic Adjustment Method for Multimodal High Grayscale Image Enhancement

Abstract

This invention relates to the fields of computer technology and image analysis technology, and particularly to an adaptive, dynamically adjustable, multimodal high grayscale image enhancement method. The method includes the following steps: First, to improve its generalization ability, an adaptive grayscale image correction mapping algorithm based on RAW prior knowledge is designed to achieve quantization processing of images with different bit depths. Second, to ensure the effectiveness of the enhanced data, an adaptive power-law compensation method is designed to compensate for the uneven brightness of the grayscale corrected image. Finally, to make the enhanced image more consistent with human visual perception, an innovative adaptive brightness and power-law correction algorithm is designed for the first time, which can more reasonably and adaptively adjust the color saturation and spatial brightness of the image. Experimental results show that this invention can achieve optimal performance on various datasets, and the processing results are more consistent with human visual perception, exhibiting strong robustness and universality.

Description

Technical Field

[0001] This invention relates to the fields of computer technology and image analysis technology, and in particular to an adaptive dynamic adjustment method for enhancing multimodal high grayscale images. Background Technology

[0002] Raw synthetic aperture radar (SAR) images, medical CT imaging, industrial X-ray inspection, and infrared images typically have a bit depth of 16 bits or higher, making them unsuitable for direct visualization on 8-bit displays. Furthermore, due to psychophysiological limitations, the human eye can only distinguish 40 grayscale levels of information, but possesses extremely high resolution for color information. Therefore, visualizing and analyzing such image data is very complex. To improve the visual resolution and contrast of grayscale images, color enhancement methods are commonly used to process black-and-white grayscale images or multi-band grayscale images into color images, thereby improving the recognizability of usable information.

[0003] Scholars have used pseudo-color technology to improve the visibility of information in low-contrast grayscale images and other image features. However, most grayscale images have more than 256 grayscale levels, while most existing pseudo-color studies are based on 256 grayscale levels. Since the dynamic range of grayscale in high-grayscale images is usually much larger than the 256 grayscale levels that an 8-bit image can represent, linear mapping or truncation based on 256 grayscale levels will compress a large amount of high-level grayscale information into the same grayscale range, resulting in loss of detail, decreased contrast, and irreversible loss of important structural features.

[0004] Based on this, a pseudo-color display method for high grayscale weld film images was proposed, with publication number "CN113643194B," and a pseudo-color enhancement method for high grayscale weld film images was designed. However, this method still has the following shortcomings in the enhancement processing of high grayscale images: 1. Existing high grayscale enhancement algorithms are prone to problems with distinct tonal levels; 2. The adaptability is weak, and the enhancement coefficients need to be manually selected, which is time-consuming and laborious; 3. Existing pseudo-color algorithms have poor universality, and their generalization ability is insufficient and their robustness is poor when processing different images generated from multiple fields. Summary of the Invention

[0005] In view of the above, in order to overcome the problems of distinct tonal gradations, weak adaptability and poor universality of existing technologies, this invention proposes an adaptive dynamic adjustment pseudo-color enhancement algorithm for multimodal high grayscale images.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: to provide an adaptive dynamic adjustment method for enhancing multimodal high grayscale images, comprising the following steps:

[0007] Step 1: Acquire multimodal images and convert the HIS color space of the grayscale image to the RGB color space for display and storage using formulas (1) and (2).

[0008] Step two: Design an adaptive grayscale image correction mapping algorithm based on RAW prior knowledge, and use this algorithm to correct the high-bit image acquired in step one. Bit depth normalization and adaptive grayscale correction mapping are performed to obtain a grayscale corrected image;

[0009] Step 3: Design an adaptive power-law correction compensation algorithm based on prior knowledge of grayscale. Use this algorithm to process the grayscale correction image input in Step 3 to obtain the adaptive power-law compensation image.

[0010] Step four: Design an adaptive brightness enhancement algorithm and a power-law correction pseudo-color enhancement algorithm based on HIS prior knowledge. Process the adaptive power-law compensation image obtained in step three using the algorithms to obtain the final pseudo-color enhancement image.

[0011] Furthermore, in step one above, the HIS color space of the grayscale image is converted to the RGB color space for display and storage using formulas (1) and (2):

[0012] (1)

[0013] (2)

[0014] in, , , This represents a pseudo-color space vector constructed from a grayscale image, containing three components: brightness, saturation, and hue. This indicates the coordinates of the input image after the previous correction steps. The grayscale value at that location. This indicates the bit depth of the image data. This represents the HIS chromatography adaptive correction factor. This represents the adaptive power-law correction compensation factor. This represents the nonlinear mapping adjustment constant, used to control the weight of the nonlinear term in the denominator. Gain coefficients representing different luminance components: Represents the luminance component The gain coefficient is used to adjust the amplitude of brightness; Represents color saturation components The enhancement factor is used to adjust the vibrancy of colors; Representing hue components The scaling factor is used to control the range of color variation. Indicates the bias adjustment coefficient: Represents the luminance component The bias compensation coefficient is used to adjust the baseline value of brightness; Represents color saturation components The bias coefficient; Representing hue components The basic rotation angle or initial phase shift determines the basic tonal tendency of the image. This represents the blue channel component that is output to the display device after conversion. This represents the green channel component that is output to the display device after conversion. This represents the red channel component that is output to the display device after conversion.

[0015] The given measurement matrix is ​​shown in formula (3).

[0016] (3)

[0017] Obtain high-bit image .

[0018] Furthermore, the specific sub-steps of step two above are as follows:

[0019] 2.1. Process the high-bit image output from step one using bit-depth normalized quantization. Obtain the normalized image ;

[0020] 2.2 An adaptive grayscale image correction mapping algorithm based on RAW prior knowledge is designed to correct the normalized image from step 2.1. As input, a grayscale corrected image is obtained.

[0021] Furthermore, in section 2.2 above, the core formula of the adaptive grayscale image correction mapping algorithm based on RAW prior knowledge is as follows:

[0022] (5)

[0023] (6)

[0024] in, and These represent the window level values ​​of the corrected image. Indicates the input image High-level grayscale statistical histogram This represents the adaptive correction coefficient introduced in this paper. Representing the image matrix respectively Width and height;

[0025] The The adaptive correction function is as follows:

[0026] (7)

[0027] Among them, input Representing the image matrix The prior mean brightness, Indicates the amplitude adjustment factor. This represents the calculated correction coefficient. As input to formula (4).

[0028] Furthermore, in step three above, the core formula (8) of the adaptive power-law correction compensation algorithm based on grayscale prior knowledge is shown.

[0029] (8)

[0030] Among them, input Representing the image matrix The prior mean brightness, This represents the prior amplitude adjustment factor. This represents the calculated power-order correction compensation coefficient.

[0031] Furthermore, the specific sub-steps of step four above are as follows:

[0032] 4.1 An adaptive brightness enhancement algorithm based on HIS prior knowledge is designed. The constructed adaptive adjustment function is used to process the adaptive power-law compensation image obtained in step three to obtain the color saturation components in the HIS color space.

[0033] 4.2 Based on the processing results of step 4.1, the spatial intensity of the low grayscale image is dynamically adjusted using the brightness distribution of the I channel in step one. The calculation formulas for the dilation coefficient and compensation factor are shown in (10) and (11), where formula (13) is the effective constraint condition for the dilation coefficient and compensation factor, which satisfies... At that time, adaptive adjustment of chromatographic brightness is performed to obtain an adaptive power-compensated image;

[0034] 4.3, Design a power-law correction pseudo-color enhancement algorithm, introduce adaptive power-law adjustment into the color spectrum of the adaptive power-law compensation image input in step 4.2, and construct formula (9). The output is used as input to dynamically and adaptively adjust the power value; then all the parameter variables introduced in formula (1) are set to adaptive dynamic adjustment based on prior knowledge to achieve adaptive enhancement of the HIS color space.

[0035] Furthermore, in section 4.1 above, the adaptive adjustment factor of the saturation component is dynamically adjusted based on the grayscale distribution of the S channel, and the calculation formula is shown in (9).

[0036] (9)

[0037] in, is the compensation coefficient for color saturation, which is set to in this paper. , This represents the value of the S channel in the HIS color space at coordinates (x, y).

[0038] Furthermore, in section 4.2 above, the formula (12) for adaptive adjustment of chromatographic brightness is as follows:

[0039] (10)

[0040] (11)

[0041] (12)

[0042] (13)

[0043] in, Control parameters for different types of images, This is a control factor.

[0044] Furthermore, in section 4.2 above, the calculation formula for the power-correction pseudo-color enhancement algorithm is shown in (14).

[0045] (14)

[0046] The calculation formulas for adaptive adjustment are as follows (15) and (16).

[0047] (15)

[0048] (16)

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. In order to improve the generalization of the model, this invention designs an adaptive grayscale image correction mapping algorithm based on RAW prior knowledge. This algorithm combines image statistical distribution features, parameter selection, and normalized bit depth constant to construct a dynamic grayscale mapping function to realize the quantization processing of images with different bit depths.

[0051] 2. To ensure the effectiveness of the augmented data, an adaptive power-law correction compensation algorithm based on grayscale prior knowledge was designed. The power-law factor is dynamically generated by calculating the global prior mean brightness of the image. Compared with existing algorithms with fixed parameters or manual adjustment, this algorithm has adaptive adjustment capability, and therefore can effectively compensate for the phenomenon of unclear brightness in grayscale corrected images.

[0052] 3. In order to make the enhanced image more in line with human visual perception, an innovative adaptive power-law correction compensation algorithm and a power-law correction pseudo-color enhancement algorithm based on grayscale prior knowledge were designed. By using the constructed adaptive adjustment function of saturation and brightness, combined with the grayscale information after power-law compensation, the color saturation and spatial brightness of the image can be more reasonably and adaptively adjusted.

[0053] 4. This invention adopts a hierarchical enhancement architecture. First, the original data is processed by adaptive grayscale correction mapping based on RAW prior knowledge to obtain a grayscale corrected image. The corrected image is then optimized a second time by an adaptive power correction compensation algorithm, which significantly enhances the local details and overall contrast of the image. Finally, the optimized data is fed into the designed HIS pseudo-color enhancement model for color mapping, and finally a color image that conforms to the visual perception characteristics of the human eye is obtained.

[0054] Experimental results show that the proposed method can achieve the best performance on a variety of datasets, and the processing results are more in line with human visual perception. It has strong robustness and universality, and also provides an effective solution for adaptive enhancement of multimodal low-quality images. Attached Figure Description

[0055] Figure 1 This is an overall flowchart of the present invention;

[0056] Figure 2 The image shows the adaptive grayscale image correction result based on RAW prior knowledge. In this image, (a) and (b) are the outputs of formula (4) and formula (5) respectively, and (c) and (d) are the statistical histograms corresponding to (a) and (b) respectively.

[0057] Figure 3 This is a schematic diagram illustrating the adaptive brightness and power-law correction effects based on HIS prior knowledge.

[0058] Figure 4 The image shows the results of high grayscale pseudo-color image processing in various fields. Rows (1) and (2) are 14-bit low-contrast FLIR infrared images, respectively; rows (3) and (4) are 14-bit medium-contrast FLIR infrared images, respectively; row (5) is a 16-bit marine remote sensing SAR image; and row (6) is a digital X-ray image of oil pipeline welding. Detailed Implementation

[0059] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments and accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] The design concept of this invention is as follows: First, to improve its generalization ability, an adaptive grayscale image correction mapping algorithm based on RAW prior knowledge is designed to achieve quantization processing of images with different bit depths. Second, to ensure the effectiveness of the enhanced data, an adaptive power-law compensation method is designed to compensate for the phenomenon of unclear brightness in the grayscale corrected image. Finally, to make the enhanced image more consistent with human visual perception, an innovative adaptive brightness and power-law correction algorithm is designed for the first time, which can more reasonably and adaptively adjust the color saturation and spatial brightness of the image.

[0061] Example: See Figure 1 The present invention provides an adaptive dynamic adjustment method for enhancing multimodal high grayscale images, which specifically includes the following steps:

[0062] Step 1: Acquire multimodal images, and convert the HIS color space of the grayscale image to the RGB color space for display and storage using formulas (1) and (2).

[0063] (1)

[0064] (2)

[0065] in: , , It represents a pseudo-color space vector constructed from a grayscale image, containing three components: brightness, color saturation, and hue. This indicates the coordinates of the input image after the previous correction steps. The grayscale value at that location; Indicates the bit depth of the image data; Indicates the HIS chromatography adaptive correction factor; This represents the adaptive power-law correction compensation factor; This represents the nonlinear mapping adjustment constant, used to control the weight of the nonlinear term in the denominator; Represents the luminance component The gain coefficient is used to adjust the amplitude of brightness; Represents the luminance component The bias compensation coefficient; Represents color saturation components The enhancement factor is used to adjust the vividness of colors. The larger the value, the higher the saturation of the output image and the more vivid the colors. Represents color saturation components The bias coefficient; Representing hue components The scaling factor is used to control the range of color variation; Representing hue components The basic rotation angle or initial phase shift determines the basic tonal tendency of the image; This represents the blue channel component output to the display device after conversion; This represents the green channel component that is output to the display device after conversion; This represents the red channel component that is output to the display device after conversion.

[0066] The given measurement matrix is ​​shown in formula (3).

[0067] (3)

[0068] Obtain high-bit image .

[0069] Step two: Design an adaptive grayscale image correction mapping algorithm based on RAW prior knowledge, and use this algorithm to correct the high-bit image acquired in step one. Perform bit depth normalization and adaptive grayscale correction mapping to obtain a grayscale corrected image: The specific sub-steps are as follows:

[0070] 2.1. Process the high-bit image output from step one using bit-depth normalized quantization. Obtain the normalized image :

[0071] The calculation formula for the bit depth normalized quantization is as follows:

[0072] (4)

[0073] in, Represents a normalized image. Indicates the output image The bit depth, and These represent the input images respectively. The maximum and minimum values.

[0074] 2.2 An adaptive grayscale image correction mapping algorithm based on RAW prior knowledge is designed to correct the normalized image from step 2.1. As input, we obtain the grayscale corrected image:

[0075] The core formula of the algorithm is as follows:

[0076] (5)

[0077] (6)

[0078] in, and These represent the window level values ​​of the corrected image. Indicates the input image High-level grayscale statistical histogram This represents the adaptive correction coefficient introduced in this paper;

[0079] The The adaptive correction function is as follows:

[0080] (7)

[0081] Among them, input Representing the image matrix The prior mean brightness, Indicates the amplitude adjustment factor. This represents the calculated correction coefficient. As input to formula (4);

[0082] The enhancement process of the infrared image using the above method is as follows: Figure 2 As shown, the low-contrast image correction process is mainly illustrated. (a) and (b) are the outputs of formulas (4) and (5), respectively, and (c) and (d) are the statistical histograms corresponding to (a) and (b), respectively. It can be seen that the image contrast is enhanced to a certain extent after adaptive grayscale image correction using RAW prior knowledge. Since the pixel sequence of the image uses the same linear mapping, the mapped 16-bit image can retain the information of the infrared image to the greatest extent.

[0083] Step 3: Design an adaptive power-law correction compensation algorithm based on prior knowledge of grayscale. This algorithm is used to process the grayscale-corrected image input in Step 3 to obtain the adaptive power-law compensated image.

[0084] The core formula (8) of the adaptive power-law correction compensation algorithm based on gray-scale prior knowledge is shown below.

[0085] (8)

[0086] Among them, input Representing the image matrix The prior mean brightness, where m and n represent the width and height of the image matrix g, respectively. This represents the prior amplitude adjustment factor. This represents the calculated power-order correction compensation coefficient.

[0087] The expected output of this invention The range is , Output At the same time, it can compensate well for low gray levels (darker areas), such as Figure 3 (b) shows the compensated image. Output At that time, it can compensate well for high gray levels (darker areas), such as Figure 3 (f) shows the compensated image.

[0088] Step four involves designing an adaptive brightness enhancement algorithm and a power-law correction pseudo-color enhancement algorithm based on HIS prior knowledge. These algorithms are then used to process the adaptive power-law compensation image obtained in step three to obtain the final pseudo-color enhanced image. The specific sub-steps are as follows:

[0089] 4.1 An adaptive brightness enhancement algorithm based on HIS prior knowledge is designed. The constructed adaptive adjustment function is used to process the adaptive power-law compensation image obtained in step three to obtain the color saturation components in the HIS color space.

[0090] The adaptive adjustment factor for the saturation component is dynamically adjusted based on the grayscale distribution of the S channel, and the calculation formula is shown in (9).

[0091] (9)

[0092] in, is the compensation coefficient for color saturation, which is set to in this paper. , This indicates the S channel in the HIS color space in coordinates. The value below;

[0093] 4.2 Based on the processing results of step 4.1, the spatial intensity of the low grayscale image is dynamically adjusted using the brightness distribution of the I channel in step one. The calculation formulas for the dilation coefficient and compensation factor are shown in (10) and (11), where formula (13) is the effective constraint condition for the dilation coefficient and compensation factor, which satisfies... At that time, adaptive adjustment of chromatographic brightness is performed, as shown in formula (12).

[0094] (10)

[0095] (11)

[0096] (12)

[0097] (13)

[0098] in, Control parameters for different types of images, As a control factor, this paper sets This indicates that images with less than 65% of the statistical mean brightness are subjected to adaptive adjustment of chromatographic brightness, and through adjustment, an adaptive power-compensated image is obtained.

[0099] 4.3, Design a power-law correction pseudo-color enhancement algorithm, introduce adaptive power-law adjustment into the color spectrum of the adaptive power-law compensation image input in step 4.2, and construct formula (9). The output is used as input to dynamically and adaptively adjust the power value. The calculation formula of the power correction pseudo-color enhancement algorithm is shown in (14).

[0100] (14)

[0101] This invention sets all the parameter variables introduced in formula (1) to be adaptively and dynamically adjusted based on prior knowledge, thus achieving adaptive enhancement of the HIS color space. The adaptively adjusted calculation formulas are (15) and (16), where This is a dynamic adjustment factor for hue, primarily used to control the final rendered hue color, allowing the image to acquire more diverse colors. This invention sets it as follows: This indicates that no hue range adjustment will be performed.

[0102] (15)

[0103] (16)

[0104] Since pseudo-color images in the HIS color space are not convenient to display directly on a computer monitor, this invention uses formula (2) to convert the HIS space to the RGB space for display, wherein the remaining parameters are set as follows: , .

[0105] The enhanced effects of the method provided by this invention are as follows: Figure 3 As shown, (a) and (e) are grayscale corrected images after RAW prior knowledge correction; (b) and (f) are images after adaptive power correction compensation processing, which can better compensate for (a) and (e), resulting in significant enhancement of local details and overall contrast of the image; (d) and (h) are the final processed results, which are color images that conform to the characteristics of human visual perception.

[0106] The method provided in this invention was implemented and tested on a computer running Windows 11, with an AMD Ryzen 5 5600H processor and Radeon Graphics 3.30 GHz, 16.0 GB of RAM, using Visual Studio 2015 and OpenCV 3.1, and programmed in C++. To verify the effectiveness of the algorithm constructed in this paper, 14-bit infrared images, 16-bit marine SAR images, and 12-bit X-ray images generated in the field of autonomous driving were analyzed from both quantitative and qualitative perspectives.

[0107] The quantitative test results are shown in Table 1.

[0108] Table 1. Quantitative experimental results of high grayscale false color in FLIR images.

[0109]

[0110] Qualitative comparative experiment results as follows Figure 4 As shown, rows (1) and (2) are 14-bit low-contrast FLIR infrared images; rows (3) and (4) are 14-bit medium-contrast FLIR infrared images; row (5) is a 16-bit marine remote sensing SAR image; and row (6) is a digital X-ray image of oil pipeline welding. The results show that this method can not only effectively enhance solid rocket motor X-ray films, but also significantly improve visual contrast and detail recognition for other modal high dynamic range, low-contrast industrial and remote sensing images.

[0111] The method of this invention has achieved the best performance in high grayscale image enhancement in multiple fields, showing state-of-the-art (SOTA) performance in both quantitative indicators and visual effects; moreover, this method can adapt to the enhancement processing of images with various brightness levels, and has strong robustness and universality.

[0112] The above description is a specific illustration of the present invention, and not a limitation thereof. Those skilled in the art can make various equivalent technical solutions without departing from the scope of the present invention; therefore, all equivalent technical solutions should be included within the protection scope of the present invention.

Claims

1. An adaptively dynamically adjusted multimodal high grayscale image enhancement method, characterized in that, Includes the following steps: Step 1: Acquire multimodal images and convert the HIS color space of the grayscale image to the RGB color space for display and storage using formulas (1) and (2). Step two: Design an adaptive grayscale image correction mapping algorithm based on RAW prior knowledge, and use this algorithm to correct the high-bit image acquired in step one. Bit depth normalization and adaptive grayscale correction mapping are performed to obtain a grayscale corrected image; Step 3: Design an adaptive power-law correction compensation algorithm based on prior knowledge of grayscale. Use this algorithm to process the grayscale correction image input in Step 3 to obtain the adaptive power-law compensation image. Step four: Design an adaptive brightness enhancement algorithm and a power-law correction pseudo-color enhancement algorithm based on HIS prior knowledge. Process the adaptive power-law compensation image obtained in step three using the algorithms to obtain the final pseudo-color enhancement image.

2. The adaptive dynamic adjustment multimodal high grayscale image enhancement method according to claim 1, characterized in that, In step one, the HIS color space of the grayscale image is converted to the RGB color space for display and storage using formulas (1) and (2): (1) (2) in: , , It represents a pseudo-color space vector constructed from a grayscale image, containing three components: brightness, color saturation, and hue. This indicates the coordinates of the input image after the previous correction steps. The grayscale value at that location; Indicates the bit depth of the image data; Indicates the HIS chromatography adaptive correction factor; This represents the adaptive power-law correction compensation factor; This represents the nonlinear mapping adjustment constant, used to control the weight of the nonlinear term in the denominator; Represents the luminance component The gain coefficient is used to adjust the amplitude of brightness; Represents the luminance component The bias compensation coefficient is used to adjust the baseline value of brightness; Represents color saturation components The enhancement factor is used to adjust the vibrancy of colors; Represents color saturation components The bias coefficient; Representing hue components The scaling factor is used to control the range of color variation; Representing hue components The basic rotation angle or initial phase shift determines the basic tonal tendency of the image; This represents the blue channel component output to the display device after conversion; This represents the green channel component that is output to the display device after conversion; This represents the red channel component output to the display device after conversion; The given measurement matrix is ​​shown in formula (3). (3) Obtain high-bit image .

3. The adaptive dynamic adjustment multimodal high grayscale image enhancement method according to claim 1, characterized in that, The specific sub-steps of step two are as follows: 2.

1. Process the high-bit image output from step one using bit-depth normalized quantization. Obtain the normalized image ; 2.2 An adaptive grayscale image correction mapping algorithm based on RAW prior knowledge is designed to correct the normalized image from step 2.

1. As input, a grayscale corrected image is obtained.

4. The adaptive dynamic adjustment multimodal high grayscale image enhancement method according to claim 3, characterized in that, In section 2.2, the core formula of the adaptive grayscale image correction mapping algorithm based on RAW prior knowledge is as follows: (5) (6) in, and These represent the window level values ​​of the corrected image. Indicates the input image High-level grayscale statistical histogram This represents the adaptive correction coefficient introduced in this paper. Let represent the width and height of the image matrix g, respectively; The The adaptive correction function is as follows: (7) Among them, input Representing the image matrix The prior mean brightness, Indicates the amplitude adjustment factor. This represents the calculated correction coefficient. As input to formula (4).

5. The adaptive dynamic adjustment method for multimodal high grayscale image enhancement according to claim 1, characterized in that, In step three, the core formula (8) of the adaptive power-law correction compensation algorithm based on gray-scale prior knowledge is shown. (8) Among them, input Representing the image matrix The prior mean brightness, Representing the image matrix respectively Width and height, This represents the prior amplitude adjustment factor. This represents the calculated power-order correction compensation coefficient.

6. The adaptive dynamic adjustment multimodal high grayscale image enhancement method according to claim 1, characterized in that, The specific sub-steps of step four are as follows: 4.1 Design an adaptive brightness enhancement algorithm based on HIS prior knowledge. Use the constructed adaptive adjustment function to process the adaptive power compensation image obtained in step 3 to obtain the color saturation components in the HIS color space. 4.2 Based on the processing results of step 4.1, the spatial intensity of the low grayscale image is dynamically adjusted using the brightness distribution of the I channel in step one. The calculation formulas for the dilation coefficient and compensation factor are shown in (10) and (11), where formula (13) is the effective constraint condition for the dilation coefficient and compensation factor, which satisfies... At that time, adaptive adjustment of chromatographic brightness is performed to obtain an adaptive power-compensated image; 4.3, Design a power-law correction pseudo-color enhancement algorithm, introduce adaptive power-law adjustment into the color spectrum of the adaptive power-law compensation image input in step 4.2, and construct formula (9). The output is used as input to dynamically and adaptively adjust the power value; Then, all the parameter variables introduced in formula (1) are set to adaptive dynamic adjustment based on prior knowledge to achieve adaptive enhancement of the HIS color space.

7. The adaptive dynamic adjustment multimodal high grayscale image enhancement method according to claim 6, characterized in that, In section 4.1, the adaptive adjustment factor of the saturation component is dynamically adjusted based on the grayscale distribution of the S channel, and the calculation formula is shown in (9). (9) in, This is the compensation coefficient for color saturation. This represents the value of the S channel in the HIS color space at coordinates (x, y).

8. The adaptive dynamic adjustment multimodal high grayscale image enhancement method according to claim 6, characterized in that, In section 4.2, the formula (12) for adaptive adjustment of chromatographic brightness is as follows: (10) (11) (12) (13) in, Control parameters for different types of images, This is a control factor.

9. The adaptive dynamic adjustment method for multimodal high grayscale image enhancement according to claim 6, characterized in that, In section 4.2, the calculation formula for the power-correction pseudo-color enhancement algorithm is shown in (14). (14) The calculation formulas for adaptive adjustment are as follows (15) and (16). (15) (16)。

Images