Breast CT image enhancement method and system

Abstract

The invention relates to the field of medical images, in particular to a mammary gland CT image enhancement method and system, and specifically, based on local structure tensor eigenvalue information and normalized local entropy information of a guide image of an original mammary gland CT image, a composite edge perception weight is obtained through calculation; performing weighted guide filtering on the original mammary gland CT image by using the composite edge perception weight to obtain a detail layer image; identifying bright detail pixels in the detail layer image and bright detail areas formed by the bright detail pixels, and calculating morphological feature parameters of the bright detail areas and bright detail feature entropies in local neighborhoods of the bright detail pixels; performing selective enhancement on the bright detail pixels based on morphological feature parameters and bright detail feature entropy to obtain an optimized detail layer image; and based on image local statistical characteristics and perception measurement based on a human vision system model, calculating a fusion weight, and according to the fusion weight, carrying out adaptive fusion on the basic layer image and the optimized detail layer image to obtain a final enhanced mammary gland CT image.

Description

Technical Field

[0001] The present invention relates to the field of medical images, and in particular to a breast CT image enhancement method and system. Background Art

[0002] Breast cancer is one of the most common malignant tumors in women worldwide. Early detection and accurate diagnosis are crucial to improving patient survival rates. Compared to traditional two-dimensional mammography, breast CT can provide tomographic images without tissue overlap, which helps to show the three-dimensional structure, morphology, and edge characteristics of lesions, especially in dense breasts. During the imaging process, breast CT images are limited by factors such as X-ray dose, detector performance, and reconstruction algorithms. The density differences between soft tissues (such as glands, fat, and fibrous tissue) are small, resulting in low overall image contrast. In particular, some early-stage tiny lesions or diffuse lesions may have very low contrast with surrounding normal tissue, making them difficult to effectively identify. In addition, to reduce radiation dose, breast CT usually uses a lower X-ray exposure, which inevitably leads to higher noise levels in the image. Especially in low-dose scanning mode, noise can seriously affect image quality and drown out tiny structures and lesion details.

[0003] Conventional image enhancement techniques, such as those based on histogram equalization, spatial filtering, transform domains, and partial differential equations, can enhance edges and details, but they can also amplify image noise or lose some grayscale information, which can affect medical judgment. Suppressing noise while preserving and sharpening edge details, especially microcalcifications, is crucial for improving the early detection and diagnostic accuracy of breast cancer. Summary of the Invention

[0004] In order to solve the above problems, in a first aspect, the present invention provides a breast CT image enhancement method, the method comprising:

[0005] Acquire an original breast CT image, calculate a composite edge-aware weight based on local structural tensor eigenvalue information and normalized local entropy information of the guided image; and perform weighted guided filtering on the original breast CT image using the composite edge-aware weight to obtain a detail layer image;

[0006] Identifying bright detail pixels in the detail layer image and the bright detail regions formed therein, calculating morphological feature parameters of the bright detail regions and bright detail feature entropy within a local neighborhood of the bright detail pixels; selectively enhancing the bright detail pixels based on the morphological feature parameters and the bright detail feature entropy to obtain an optimized detail layer image;

[0007] A fusion weight is calculated based on local statistical characteristics of the image and a perceptual metric based on a human visual system model, and the base layer image and the optimized detail layer image are adaptively fused according to the fusion weight to obtain a final enhanced breast CT image.

[0008] Preferably, the composite edge perception weight is calculated based on the local structure tensor eigenvalue information and the normalized local entropy information of the guidance image, specifically:

[0009] Calculate the structure tensor of the guidance image in each pixel neighborhood and solve its eigenvalue;

[0010] Calculate the normalized local entropy of the guidance image within each pixel neighborhood;

[0011] The composite edge-aware weight is calculated based on the eigenvalue and the normalized local entropy.

[0012] Preferably, the selective enhancement of bright detail pixels based on the morphological feature parameters and bright detail feature entropy is specifically performed as follows:

[0013] Obtain pixels whose brightness is greater than the brightness threshold of bright detail pixels and perform connected domain analysis to obtain bright detail areas;

[0014] Calculating morphological characteristic parameters of each bright detail area, wherein the parameters include at least one of area, circularity, and density;

[0015] The local bright detail feature entropy of each bright detail pixel is calculated to obtain a target range of the morphological feature parameter and a target interval of the entropy value; when the morphological feature parameter of the region to which the pixel belongs is within the target range and the bright detail feature entropy is within the target interval, a first enhancement rule is applied to the pixel; otherwise, a second enhancement rule is applied, and the enhancement strength corresponding to the first enhancement rule is greater than that of the second enhancement rule.

[0016] Preferably, the fusion weights used in the adaptive fusion process include base layer weights and detail layer weights;

[0017] Calculating the local variance of the base layer image at pixel i and a perceptual metric based on a human visual system model, the metric comprising at least one of local contrast sensitivity or visual masking effect;

[0018] The detail layer weight is calculated using a preset rule according to the local variance, the strength of the optimized detail layer value, and the perception metric.

[0019] Preferably, the method further comprises suppressing non-bright detail pixels in the detail layer image.

[0020] In a second aspect of the present invention, a breast CT image enhancement system is provided, the system comprising:

[0021] A guided filtering module is configured to obtain an original breast CT image, calculate a composite edge-aware weight based on local structural tensor eigenvalue information and normalized local entropy information of the guided image, and perform weighted guided filtering on the original breast CT image using the composite edge-aware weight to obtain a detail layer image;

[0022] a detail enhancement module configured to identify bright detail pixels in the detail layer image and the bright detail regions formed therein, calculate morphological feature parameters of the bright detail regions, and bright detail feature entropy within a local neighborhood of the bright detail pixels; and selectively enhance the bright detail pixels based on the morphological feature parameters and the bright detail feature entropy to obtain an optimized detail layer image;

[0023] A fusion module is used to calculate a fusion weight based on local statistical characteristics of the image and a perceptual metric based on a human visual system model, and adaptively fuse the base layer image and the optimized detail layer image according to the fusion weight to obtain a final enhanced breast CT image.

[0024] Preferably, the composite edge perception weight is calculated based on the local structure tensor eigenvalue information and the normalized local entropy information of the guidance image, specifically:

[0025] Calculate the structure tensor of the guidance image in each pixel neighborhood and solve its eigenvalue;

[0026] Calculate the normalized local entropy of the guidance image within each pixel neighborhood;

[0027] The composite edge-aware weight is calculated based on the eigenvalue and the normalized local entropy.

[0028] Preferably, the selective enhancement of bright detail pixels based on the morphological feature parameters and bright detail feature entropy is specifically performed as follows:

[0029] Obtain pixels whose brightness is greater than the brightness threshold of bright detail pixels and perform connected domain analysis to obtain bright detail areas;

[0030] Calculating morphological characteristic parameters of each bright detail area, wherein the parameters include at least one of area, circularity, and density;

[0031] The local bright detail feature entropy of each bright detail pixel is calculated to obtain a target range of the morphological feature parameter and a target interval of the entropy value; when the morphological feature parameter of the region to which the pixel belongs is within the target range and the bright detail feature entropy is within the target interval, a first enhancement rule is applied to the pixel; otherwise, a second enhancement rule is applied, and the enhancement strength corresponding to the first enhancement rule is greater than that of the second enhancement rule.

[0032] Preferably, the fusion weights used in the adaptive fusion process include base layer weights and detail layer weights;

[0033] Calculating the local variance of the base layer image at pixel i and a perceptual metric based on a human visual system model, the metric comprising at least one of local contrast sensitivity or visual masking effect;

[0034] The detail layer weight is calculated using a preset rule according to the local variance, the strength of the optimized detail layer value, and the perception metric.

[0035] Finally, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in the first aspect.

[0036] The present invention employs weighted guided filtering using composite edge-aware weights that combine the eigenvalues of the structural tensor with local entropy. This allows for more accurate differentiation between true edges, texture, and noise, effectively smoothing noise while more precisely preserving and even sharpening tissue boundaries and lesion outlines in the image. Furthermore, by combining morphological analysis of bright details with entropy analysis of bright detail features, the method achieves highly selective enhancement of bright details with specific morphologies and low entropy characteristics, while simultaneously suppressing noise and irrelevant texture, improving the signal-to-noise ratio of critical diagnostic information. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of Example 1;

[0038] Figure 2 Schematic diagram of the original image, base layer, and detail layer;

[0039] Figure 3 Optimized before and after comparison images for detail layers;

[0040] Figure 4 Before and after optimization of the original image;

[0041] Figure 5 This is a structural diagram of Example 2. DETAILED DESCRIPTION

[0042] In this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0044] Example 1, as Figure 1 The breast CT image enhancement method shown includes:

[0045] S1, obtaining an original breast CT image, calculating a composite edge-aware weight based on local structural tensor eigenvalue information and normalized local entropy information of a guided image; and performing weighted guided filtering on the original breast CT image using the composite edge-aware weight to obtain a detail layer image;

[0046] Obtain the original breast CT image I to be processed, set the guide image G, preferably G = I. Calculate the composite edge perception weight F * In one embodiment, the composite edge-aware weight is calculated based on the local structure tensor eigenvalue information and the normalized local entropy information of the guidance image, specifically:

[0047] Calculate the structure tensor of the guidance image in each pixel neighborhood and solve its eigenvalue;

[0048] Calculate the normalized local entropy of the guidance image within each pixel neighborhood;

[0049] The composite edge-aware weight is calculated based on the eigenvalue and the normalized local entropy.

[0050] For the guidance image G, calculate its structure tensor j in the neighborhood of each pixel i iThe structure tensor is a matrix constructed based on the image gradient information within the pixel neighborhood, which can reflect the structural pattern of the local image. Solve the eigenvalue λ of the structure tensor j,1 and λ j,2 ,λ j,1 ≥λ j,2 If both eigenvalues are small, it indicates a flat region. If λ j,1 Much larger than λ j,2 , indicating strong edges or linear structures; larger values of both indicate corners or complex textures.

[0051] Furthermore, the normalized local entropy H of the guidance image G in the neighborhood of each pixel i is calculated norm,i The local entropy represents the randomness or complexity of the grayscale distribution within the pixel neighborhood. A high entropy value corresponds to an area with rich texture or more noise, while a low entropy value corresponds to an area with a single or flat grayscale. j,1 and λ j,2 and normalized local entropy H norm,i Calculating composite edge-aware weights In one embodiment, when the eigenvalues indicate the presence of a strong and simple edge structure, such as λ j,1 Very large and much larger than λ j,2 , and the local entropy H norm,i When is low, a higher weight value is assigned; when the eigenvalue indicates a flat area, such as λ j,1 ,λ j,2 The average value is small or the eigenvalue and entropy value together indicate the noise / complex texture area, such as H norm,i When the value is very high, a lower weight is assigned. This invention combines geometric structure and information complexity in its weight calculation method. Compared with traditional methods that rely solely on variance, it can more robustly and accurately reflect the edge preservation requirements at the pixel location. This overcomes the shortcomings of traditional weights that are based solely on the intensity of brightness changes, which are susceptible to noise interference and difficult to distinguish between texture and edges.

[0052] For example, if the pixel is located on a clear tissue boundary, the calculated eigenvalue is λ j,1 =100,λ j,2 =5, entropy H norm,i = 0.1, these values indicate strong edges, simple structures, and the calculated composite weights will be very high, for example close to 1. If the pixel is located inside the uniform fat tissue, the eigenvalue is λ j,1 =2,λ j,2 =1, entropy H norm,i = 0.05, indicating a flat area, the calculated weight If the pixel is located in a noisy area, the eigenvalue may be uncertain, but the entropy H norm,i = 0.8 will be very high, and high entropy will result in the calculated weights Very low, so that different processing strengths can be applied to different areas during subsequent filtering.

[0053] Using the calculated composite edge-aware weight Γ * , perform weighted guided filtering operation on the original breast CT image I. It should be noted that, in the present invention, the weighted guided filtering directly outputs a smoothed version of the image, namely the base layer image B. The base layer B uses Γ * Weighted, it can effectively smooth the noise while maintaining the good * The important edge structures are identified. Then, the detail layer image D = IB is obtained by performing a difference operation between the original image I and the base layer B, as shown in Figure 2 As shown in Figure 2, the detail layer D contains high-frequency information separated from the original image, including noise and various fine structures.

[0054] S2, identifying bright detail pixels in the detail layer image and the bright detail regions formed therein, calculating morphological feature parameters of the bright detail regions and bright detail feature entropy within a local neighborhood of the bright detail pixels; selectively enhancing the bright detail pixels based on the morphological feature parameters and the bright detail feature entropy to obtain an optimized detail layer image;

[0055] Identify all brightness values in D that are greater than the brightness threshold T bright Pixels with bright detail pixels are labeled as bright detail pixels. Connected domain analysis is performed on these bright detail pixels to identify regions consisting of interconnected bright pixels, i.e., bright detail regions. For each identified bright detail region, its morphological characteristic parameters are calculated, such as its area, circularity, and density. Density is the ratio of the average brightness of the pixels in the region to the maximum brightness, or the ratio of its area to the area of the bounding rectangle.

[0056] For each bright detail pixel i, calculate the local bright detail feature entropy H of the brightness value distribution of other bright detail pixels in the local neighborhood (such as 3x3 or 5x5 window) BDFE,i , low entropy indicates high brightness consistency of bright pixels in the neighborhood. Then, selective enhancement is performed, and the target range of morphological parameters (such as the typical area range corresponding to microcalcification points, the lower limit of circularity, and the lower limit of density) and the target interval of entropy value (such as the low entropy interval) are pre-set. For each bright detail pixel i: determine whether the morphological parameters of the area to which it belongs fall within the target range, and whether its own H BDFE,i If both of them are satisfied, that is, the morphology is similar to microcalcification and the interior is homogeneous, then the pixel is considered to be a feature that needs attention, and the first enhancement rule is applied to it, that is, strong enhancement is performed, for example, its brightness value is multiplied by a large enhancement coefficient k strong>1. If at least one of the following conditions is not met, i.e., the shape is irregular, the size is inconsistent, or the internal brightness changes greatly, then it is considered to be more likely to be noise or ordinary texture, and a second enhancement method is applied to it, i.e., weak enhancement or no enhancement, for example, multiplying it by a smaller coefficient k weak , preferably, 1≤k weak <k strong , or k weak = 1. In an optional embodiment, for non-bright detail pixels in the detail layer D, that is, pixels with brightness values lower than T bright , then apply suppression processing, such as multiplying its brightness value by a suppression factor less than 1, or performing threshold processing to remove weak noise. After completing the above operations, the optimized detail layer image D′ is obtained, as shown in Figure 3 shown.

[0057] S3, calculating a fusion weight based on local statistical characteristics of the image and a perceptual metric based on a human visual system model, and adaptively fusing the base layer image and the optimized detail layer image according to the fusion weight to obtain a final enhanced breast CT image.

[0058] Calculate the local statistical characteristics of the image required for fusion. In another embodiment, the local statistical characteristics of the image are the local variance of the base layer B at each pixel i, and / or the intensity information of the optimized detail layer D′ at pixel i, such as its absolute value |D′| or the value after function mapping, and then calculate the perceptual metric based on the HVS model. The perceptual metric includes but is not limited to local contrast sensitivity, visual masking effect index, etc. In another embodiment, the contrast sensitivity is based on the spatial frequency characteristics of the neighborhood of pixel i, estimating the sensitivity of the human eye to the contrast change at that location, and the visual masking effect index is based on the texture complexity or edge strength of the neighborhood of pixel i, estimating the degree of masking of detail information by the strong background.

[0059] Calculate the detail layer fusion weight β based on the local statistical characteristics and HVS perception metrics i When the local variance of the base layer is large and the intensity of the optimized detail layer is high, increasing β i If the contrast sensitivity of the human eye is high and the masking effect is weak, maintain or further increase β i On the contrary, if the sensitivity is low or the masking effect is strong, β will be significantly reduced. i This is because in the latter case, injecting details may not contribute to perceptual improvement but may introduce visual noise. i After that, the base layer weight is α i =1-β i According to I enhanced (i) = α i B i +βi D′ i Get the final enhanced image I enhanced (i), such as Figure 4 shown.

[0060] Example 2, as Figure 5 As shown, a breast CT image enhancement system is provided, the system comprising:

[0061] A guided filtering module is configured to obtain an original breast CT image, calculate a composite edge-aware weight based on local structural tensor eigenvalue information and normalized local entropy information of the guided image, and perform weighted guided filtering on the original breast CT image using the composite edge-aware weight to obtain a detail layer image;

[0062] a detail enhancement module configured to identify bright detail pixels in the detail layer image and the bright detail regions formed therein, calculate morphological feature parameters of the bright detail regions, and bright detail feature entropy within a local neighborhood of the bright detail pixels; and selectively enhance the bright detail pixels based on the morphological feature parameters and the bright detail feature entropy to obtain an optimized detail layer image;

[0063] A fusion module is used to calculate a fusion weight based on local statistical characteristics of the image and a perceptual metric based on a human visual system model, and adaptively fuse the base layer image and the optimized detail layer image according to the fusion weight to obtain a final enhanced breast CT image.

[0064] Preferably, the composite edge perception weight is calculated based on the local structure tensor eigenvalue information and the normalized local entropy information of the guidance image, specifically:

[0065] Calculate the structure tensor of the guidance image in each pixel neighborhood and solve its eigenvalue;

[0066] Calculate the normalized local entropy of the guidance image within each pixel neighborhood;

[0067] The composite edge-aware weight is calculated based on the eigenvalue and the normalized local entropy.

[0068] Preferably, the selective enhancement of bright detail pixels based on the morphological feature parameters and bright detail feature entropy is specifically performed as follows:

[0069] Obtain pixels whose brightness is greater than the brightness threshold of bright detail pixels and perform connected domain analysis to obtain bright detail areas;

[0070] Calculating morphological characteristic parameters of each bright detail area, wherein the parameters include at least one of area, circularity, and density;

[0071] The local bright detail feature entropy of each bright detail pixel is calculated to obtain a target range of the morphological feature parameter and a target interval of the entropy value; when the morphological feature parameter of the region to which the pixel belongs is within the target range and the bright detail feature entropy is within the target interval, a first enhancement rule is applied to the pixel; otherwise, a second enhancement rule is applied, and the enhancement strength corresponding to the first enhancement rule is greater than that of the second enhancement rule.

[0072] Preferably, the fusion weights used in the adaptive fusion process include base layer weights and detail layer weights;

[0073] Calculating the local variance of the base layer image at pixel i and a perceptual metric based on a human visual system model, the metric comprising at least one of local contrast sensitivity or visual masking effect;

[0074] The detail layer weight is calculated using a preset rule according to the local variance, the strength of the optimized detail layer value, and the perception metric.

[0075] Embodiment 3 provides a computer-readable storage medium, on which a computer program is stored. The computer program, when executed by a processor, implements the method described in embodiment 1.

[0076] Through the description of the above embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by adding the necessary general hardware platform, or of course, by combining hardware and software. Based on this understanding, the essence of the above technical solution or the portion that contributes to the prior art can be embodied in the form of a computer product. The present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it, and other embodiments may also be used. Although the present invention has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A breast CT image enhancement method, characterized in that: The method comprises: Acquire an original breast CT image, calculate a composite edge-aware weight based on local structural tensor eigenvalue information and normalized local entropy information of the guided image; and perform weighted guided filtering on the original breast CT image using the composite edge-aware weight to obtain a detail layer image; Identifying bright detail pixels in the detail layer image and the bright detail regions formed therein, calculating morphological feature parameters of the bright detail regions and bright detail feature entropy within a local neighborhood of the bright detail pixels; selectively enhancing the bright detail pixels based on the morphological feature parameters and the bright detail feature entropy to obtain an optimized detail layer image; A fusion weight is calculated based on local statistical characteristics of the image and a perceptual metric based on a human visual system model, and the base layer image and the optimized detail layer image are adaptively fused according to the fusion weight to obtain a final enhanced breast CT image.

2. The method according to claim 1, wherein The composite edge perception weight is calculated based on the local structure tensor eigenvalue information and the normalized local entropy information of the guidance image, specifically: Calculate the structure tensor of the guidance image in each pixel neighborhood and solve its eigenvalue; Calculate the normalized local entropy of the guidance image within each pixel neighborhood; The composite edge-aware weight is calculated based on the eigenvalue and the normalized local entropy.

3. The method according to claim 1, wherein The selective enhancement of bright detail pixels based on the morphological feature parameters and bright detail feature entropy is specifically as follows: Obtain pixels whose brightness is greater than the brightness threshold of bright detail pixels and perform connected domain analysis to obtain bright detail areas; Calculating morphological characteristic parameters of each bright detail area, wherein the parameters include at least one of area, circularity, and density; The local bright detail feature entropy of each bright detail pixel is calculated to obtain a target range of the morphological feature parameter and a target interval of the entropy value; when the morphological feature parameter of the region to which the pixel belongs is within the target range and the bright detail feature entropy is within the target interval, a first enhancement rule is applied to the pixel; otherwise, a second enhancement rule is applied, and the enhancement strength corresponding to the first enhancement rule is greater than that of the second enhancement rule.

4. The method according to claim 1, wherein The fusion weights used in the adaptive fusion process include base layer weights and detail layer weights; Calculating the local variance of the base layer image at pixel i and a perceptual metric based on a human visual system model, the metric comprising at least one of local contrast sensitivity or visual masking effect; The detail layer weight is calculated using a preset rule according to the local variance, the strength of the optimized detail layer value, and the perception metric.

5. The method according to claim 1, wherein The method further includes suppressing non-bright detail pixels in the detail layer image.

6. A breast CT image enhancement system, characterized in that: The system comprises: A guided filtering module is configured to obtain an original breast CT image, calculate a composite edge-aware weight based on local structural tensor eigenvalue information and normalized local entropy information of the guided image, and perform weighted guided filtering on the original breast CT image using the composite edge-aware weight to obtain a detail layer image; a detail enhancement module configured to identify bright detail pixels in the detail layer image and the bright detail regions formed therein, calculate morphological feature parameters of the bright detail regions, and bright detail feature entropy within a local neighborhood of the bright detail pixels; and selectively enhance the bright detail pixels based on the morphological feature parameters and the bright detail feature entropy to obtain an optimized detail layer image; A fusion module is used to calculate a fusion weight based on local statistical characteristics of the image and a perceptual metric based on a human visual system model, and adaptively fuse the base layer image and the optimized detail layer image according to the fusion weight to obtain a final enhanced breast CT image.

7. The system according to claim 6, wherein: The composite edge perception weight is calculated based on the local structure tensor eigenvalue information and the normalized local entropy information of the guidance image, specifically: Calculate the structure tensor of the guidance image in each pixel neighborhood and solve its eigenvalue; Calculate the normalized local entropy of the guidance image within each pixel neighborhood; The composite edge-aware weight is calculated based on the eigenvalue and the normalized local entropy.

8. The system according to claim 6, wherein: The selective enhancement of bright detail pixels based on the morphological feature parameters and bright detail feature entropy is specifically as follows: Obtain pixels whose brightness is greater than the brightness threshold of bright detail pixels and perform connected domain analysis to obtain bright detail areas; Calculating morphological characteristic parameters of each bright detail area, wherein the parameters include at least one of area, circularity, and density; The local bright detail feature entropy of each bright detail pixel is calculated to obtain a target range of the morphological feature parameter and a target interval of the entropy value; when the morphological feature parameter of the region to which the pixel belongs is within the target range and the bright detail feature entropy is within the target interval, a first enhancement rule is applied to the pixel; otherwise, a second enhancement rule is applied, and the enhancement strength corresponding to the first enhancement rule is greater than that of the second enhancement rule.

9. The system according to claim 6, wherein: The fusion weights used in the adaptive fusion process include base layer weights and detail layer weights; Calculating the local variance of the base layer image at pixel i and a perceptual metric based on a human visual system model, the metric comprising at least one of local contrast sensitivity or visual masking effect; The detail layer weight is calculated using a preset rule according to the local variance, the strength of the optimized detail layer value, and the perception metric.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 5.

Images