Real-time endoscopic image detail enhancement method and device

Abstract

The present invention discloses a real-time endoscopic image detail enhancement method and device, belonging to the field of image processing. The real-time endoscopic image detail enhancement method of the present invention comprises the following steps: S1, obtaining an original endoscopic image; S2, performing guided filtering on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image; S3, performing mask sharpening processing on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail enhanced image. The present invention can quickly and effectively improve the quality of endoscopic images.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and more specifically, to a real-time endoscopic image detail enhancement method and device. Background Art

[0002] In clinical surgery using endoscopes, high-quality and low-latency endoscopic images are important conditions for ensuring the success of the operation. Affected by factors such as image sensor resolution, lighting conditions, internal tissue environment, and light source characteristics, it is easy to have problems such as missing image details such as local tissues and blood vessels, which can affect the doctor's judgment and increase the risk of surgery.

[0003] For disposable endoscopes, due to strict cost constraints, the resolution and dynamic range of image sensors are often limited. To ensure efficient target recognition performance, it is even more necessary to perform real-time detail enhancement on endoscopic images.

[0004] Current endoscopic image detail enhancement schemes are generally divided into two categories. One is the traditional sharpening filtering method, which has the advantages of low computational complexity and good real-time performance, but it is easy to increase image noise and lose local detail characteristics; the other is the deep learning method, which has better image noise control and detail retention effects, but has high computational complexity and large processing delay, and can only be applied to diagnostic scenarios with a high tolerance for delay. Summary of the invention

[0005] 1. Technical problem to be solved by the invention

[0006] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a real-time endoscopic image detail enhancement method and device with low computational complexity, good real-time performance, and the ability to highlight the details and edges of the image and retain the overall brightness and color information of the image, thereby being able to quickly and effectively improve the quality of the endoscopic image.

[0007] 2. Technical solution

[0008] In order to achieve the above object, the technical solution provided by the present invention is:

[0009] The real-time endoscopic image detail enhancement method of the present invention comprises the following steps: S1, acquiring an original endoscopic image; S2, performing guided filtering on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image; S3, performing mask sharpening processing on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail enhanced image.

[0010] As a further improvement of the present invention, step S2 specifically includes:

[0011] S21, segmenting the original endoscopic image through a two-dimensional window set;

[0012] S22, calculating the mean and variance of the pixels in each window segmentation area;

[0013] S23, obtaining edge features corresponding to each window by the following formula, and calculating edge perception factors of each window according to the edge features:

[0014]

[0015] in, represents the edge feature corresponding to the kth window, represents the local variance of the k-th window, G(·) represents the Gaussian filter, |grad| represents the absolute value of the gradient, Γ(k) represents the edge perception factor of the k-th window, N is the total number of pixels of the original endoscopic image, and λ is the edge adjustment factor;

[0016] S24, according to the mean and variance of the pixels in each window segmentation area and the edge perception factor of each window, the guidance weight corresponding to each window is calculated by the following formula:

[0017]

[0018] Among them, a k and b k is the guidance weight corresponding to the kth window, σ 2 k is the variance of the pixels in the kth window segmentation area, ε is the regularization factor, γ k is the detail adjustment factor, u k is the mean value of the pixels in the k-th window segmentation area;

[0019] S25, according to the guidance weight corresponding to each window, the base layer image is calculated by the following formula:

[0020]

[0021] Among them, i is the serial number of the pixel point, k:i∈ω k represents the kth window segmentation area including pixel point i, |ω| represents the total number of windows including pixel i, I represents the original endoscopic image, and Q represents the base layer image.

[0022] Step S3 specifically includes:

[0023] S31, performing pixel-by-pixel subtraction between the original endoscopic image and the base layer image to obtain a detail layer image;

[0024] S32, weighted synthesis of the original endoscopic image and the detail layer image by the following formula to obtain a detail enhanced image:

[0025] R i =I i +η k ·D i

[0026] Wherein, R represents the detail enhanced image, D represents the detail layer image, and n k Indicates the sharpening factor of the k-th window segmentation area where the i-th pixel is located.

[0027] The real-time endoscopic image detail enhancement method further includes: adaptively updating the edge adjustment factor, the detail adjustment factor and the sharpening factor through an adaptive parameter updating algorithm.

[0028] The edge adjustment factor is updated by the following formula:

[0029] λ=(0.001×L) 2 Among them, L is the dynamic range of each window segmentation area;

[0030] The detail adjustment factor is updated by the following formula:

[0031]

[0032] Among them, x is 0.3 times the mean value of the pixels in each window segmentation area, and m is calculated by the following formula:

[0033]

[0034] The sharpening factor is updated by:

[0035] η k =γ k +4.5.

[0036] The real-time endoscopic image detail enhancement device of the present invention comprises: an acquisition module, the acquisition module is used to acquire the original endoscopic image; a filtering module, the filtering module is used to perform guided filtering on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image; a sharpening module, the sharpening module is used to perform mask sharpening processing on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail enhanced image.

[0037] As a further improvement of the present invention, the filtering module is specifically used for:

[0038] Segmenting the original endoscopic image through a two-dimensional window set;

[0039] Calculate the mean and variance of the pixels in each window segmentation area;

[0040] The edge features corresponding to each window are obtained by the following formula, and the edge perception factor of each window is calculated according to the edge features:

[0041]

[0042] in, represents the edge feature corresponding to the kth window, represents the local variance of the k-th window, G(·) represents the Gaussian filter, |grad| represents the absolute value of the gradient, Γ(k) represents the edge perception factor of the k-th window, N is the total number of pixels of the original endoscopic image, and λ is the edge adjustment factor;

[0043] According to the mean and variance of the pixels in each window segmentation area and the edge perception factor of each window, the guidance weight corresponding to each window is calculated by the following formula:

[0044]

[0045] Among them, a k and b k is the guidance weight corresponding to the kth window, σ 2 k is the variance of the pixels in the kth window segmentation area, ε is the regularization factor, γ k is the detail adjustment factor, u k is the mean value of the pixels in the k-th window segmentation area;

[0046] According to the guidance weight corresponding to each window, the base layer image is calculated by the following formula:

[0047]

[0048] Among them, i is the serial number of the pixel point, k:i∈ω k represents the kth window segmentation area including pixel point i, |ω| represents the total number of windows including pixel i, I represents the original endoscopic image, and Q represents the base layer image.

[0049] The sharpening module is specifically used for:

[0050] Subtracting the original endoscopic image from the base layer image pixel by pixel to obtain a detail layer image;

[0051] The original endoscopic image and the detail layer image are weighted and synthesized by the following formula to obtain a detail enhanced image:

[0052] Ri =I i +η k ·D i

[0053] Wherein, R represents the detail enhanced image, D represents the detail layer image, and n k Indicates the sharpening factor of the k-th window segmentation area where the i-th pixel is located.

[0054] The real-time endoscopic image detail enhancement device further includes: an updating module, which is used to adaptively update the edge adjustment factor, the detail adjustment factor and the sharpening factor through an adaptive parameter updating algorithm.

[0055] The updating module updates the edge adjustment factor by the following formula:

[0056] λ=(0.001×L) 2

[0057] Among them, L is the dynamic range of each window segmentation area;

[0058] The updating module updates the detail adjustment factor by the following formula:

[0059]

[0060] Among them, x is 0.3 times the mean value of the pixels in each window segmentation area, and m is calculated by the following formula:

[0061]

[0062] The updating module updates the sharpening factor by the following formula:

[0063] η k =γ k +4.5.

[0064] 3. Beneficial effects

[0065] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0066] The present invention performs guided filtering on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image, and performs mask sharpening processing on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail enhanced image. The present invention has low computational complexity and good real-time performance, and can highlight the details and edges of the image and retain the overall brightness and color information of the image, thereby being able to quickly and effectively improve the quality of the endoscopic image. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 is a flow chart of a method for real-time endoscopic image detail enhancement according to an embodiment of the present invention;

[0068] FIG2( a ) is an original endoscopic image according to a specific embodiment of the present invention;

[0069] Figure 2(b) is the base layer image obtained after guiding filtering of Figure 2(a);

[0070] Figure 2(c) is a detail-enhanced image obtained after performing mask sharpening processing on Figure 2(b);

[0071] Figure 3 4 is a block diagram of a real-time endoscopic image detail enhancement device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0072] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments.

[0073] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention, so they have no substantial technical significance. Any modification of the structure, change of the proportion relationship or adjustment of the size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical content disclosed by the present invention. At the same time, the terms such as "upper", "lower", "left", "right", "middle" and so on quoted in this specification are only for the convenience of description, and are not used to limit the scope of implementation. The change or adjustment of their relative relationship should also be regarded as the scope of implementation of the present invention without substantially changing the technical content.

[0074] The present invention will be further described below in conjunction with the embodiments.

[0075] like Figure 1 As shown, the real-time endoscopic image detail enhancement method of the embodiment of the present invention includes the following steps S1 to S3:

[0076] S1, obtain the original endoscopic image.

[0077] S2, performing guided filtering on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image.

[0078] S3, performing mask sharpening processing on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail enhanced image.

[0079] In one embodiment of the present invention, the original endoscopic image may be acquired in real time by the endoscope during surgery.

[0080] In one embodiment of the present invention, step S2 specifically includes S21 to S25:

[0081] S21, through the two-dimensional window set {ω k}Segment the original endoscopic image.

[0082] S22, calculating the mean and variance of the pixels in each window segmentation area.

[0083] S23, obtaining the edge features corresponding to each window by the following formula, and calculating the edge perception factor of each window according to the edge features:

[0084]

[0085] in, represents the edge feature corresponding to the kth window, represents the local variance of the kth window, G(·) represents the Gaussian filter, |grad| represents the absolute value of the gradient, Γ(k) represents the edge perception factor of the kth window, N is the total number of pixels of the original endoscopic image, and λ is the edge adjustment factor.

[0086] S24, according to the mean and variance of the pixels in each window segmentation area and the edge perception factor of each window, the guidance weight corresponding to each window is calculated by the following formula:

[0087]

[0088] Among them, a k and b k is the guidance weight corresponding to the kth window, σ 2 k is the variance of the pixels in the kth window segmentation area, ε is the regularization factor, γ k is the detail adjustment factor, u k is the mean value of the pixels in the k-th window segmentation area.

[0089] S25, according to the guidance weight corresponding to each window, the base layer image is calculated by the following formula:

[0090]

[0091] Among them, i is the serial number of the pixel point, k:i∈ω k represents the kth window segmentation area containing pixel point i, |ω| represents the total number of windows containing pixel i, and Q represents the base layer image. This formula means that when multiple windows overlap and the segmentation areas of the multiple windows all contain pixel point i, the above-mentioned summation is performed on the pixel of the multiple windows and the average is taken.

[0092] In one embodiment of the present invention, step S3 specifically includes S31-S32:

[0093] S31, subtract the original endoscopic image from the base layer image pixel by pixel to obtain a detail layer image, and the calculation formula is as follows:

[0094] D i =I i -Q i

[0095] Where D represents the detail layer image.

[0096] S32, weighted synthesis of the original endoscopic image and the detail layer image is performed by the following formula to obtain a detail enhanced image:

[0097] R i =I i +η k ·D i

[0098] Where R represents the detail enhanced image, η k Indicates the sharpening factor of the kth window segmentation area where the i-th pixel is located. The sharpening factor corresponding to each window is the same, and all pixels in each window segmentation area are calculated using the corresponding sharpening factor.

[0099] Furthermore, the real-time endoscopic image detail enhancement method of the embodiment of the present invention may also include: adaptively updating the edge adjustment factor, the detail adjustment factor and the sharpening factor through an adaptive parameter updating algorithm.

[0100] Specifically, the edge adjustment factor can be updated by the following formula:

[0101] λ=(0.001×L) 2

[0102] Among them, L is the dynamic range of each window segmentation area.

[0103] The detail adjustment factor can be updated by:

[0104]

[0105] Among them, x is 0.3 times the mean value of the pixels in each window segmentation area, and m is calculated by the following formula:

[0106]

[0107] The sharpening factor can be updated by:

[0108] η k =γ k +4.5.

[0109] In a specific embodiment of the present invention, the original endoscopic image is shown in FIG2(a), which is an RGB image with a resolution of 548*548. The guided filter uses a circular window set with a radius of 8 pixels. According to the processing flow of step S2 above, the base layer image shown in FIG2(b) is obtained. Then, according to the processing flow of step S3 above, the detail enhanced image shown in FIG2(c) is obtained. It can be seen that after the detail enhancement is performed using the real-time endoscopic image detail enhancement method of the embodiment of the present invention, the subtle texture and edge features in the endoscopic image are significantly enhanced, the blood vessel texture becomes very clear, and even the tiny details inside the blood vessel can be seen, and the blood vessel color becomes more vivid; at the same time, the bright spot edge in the image presents a clearer and sharper outline, and the bright spot is fuller and brighter, which greatly improves the quality of the endoscopic image.

[0110] According to the real-time endoscopic image detail enhancement method of an embodiment of the present invention, guided filtering is performed on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image, and mask sharpening processing is performed on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail enhanced image. The method has low computational complexity and good real-time performance, and can highlight the details and edges of the image and retain the overall brightness and color information of the image, thereby being able to quickly and effectively improve the quality of the endoscopic image.

[0111] Corresponding to the real-time endoscopic image detail enhancement method of the above embodiment, the present invention also proposes a real-time endoscopic image detail enhancement device.

[0112] like Figure 3 As shown, the real-time endoscopic image detail enhancement device of the embodiment of the present invention comprises an acquisition module 10, a filtering module 20 and a sharpening module 30. The acquisition module 10 is used to acquire the original endoscopic image; the filtering module 20 is used to perform guided filtering on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image; the sharpening module 30 is used to perform mask sharpening processing on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail enhanced image.

[0113] In one embodiment of the present invention, the original endoscopic image may be acquired by the endoscope in real time during the operation.

[0114] In one embodiment of the present invention, the filtering module 20 first passes through a two-dimensional window set {ω k The original endoscopic image is segmented, and the mean and variance of the pixels in each window segmentation area are calculated. The edge features corresponding to each window are obtained by the following formula, and the edge perception factor of each window is calculated based on the edge features:

[0115]

[0116] in, represents the edge feature corresponding to the kth window, represents the local variance of the kth window, G(·) represents the Gaussian filter, |grad| represents the absolute value of the gradient, Γ(k) represents the edge perception factor of the kth window, N is the total number of pixels of the original endoscopic image, and λ is the edge adjustment factor.

[0117] Then, the filtering module 20 calculates the guidance weight corresponding to each window according to the mean and variance of the pixels in each window segmentation area and the edge perception factor of each window by the following formula:

[0118]

[0119] Among them, a k and b k is the guidance weight corresponding to the kth window, σ 2 k is the variance of the pixels in the kth window segmentation area, ε is the regularization factor, γ k is the detail adjustment factor, u k is the mean value of the pixels in the k-th window segmentation area.

[0120] Finally, the filtering module 20 calculates the base layer image according to the guidance weight corresponding to each window through the following formula:

[0121]

[0122] Among them, i is the serial number of the pixel point, k:i∈ω k represents the kth window segmentation area containing pixel point i, |ω| represents the total number of windows containing pixel i, I represents the original endoscopic image, Q represents the base layer image, and this formula indicates that when multiple windows overlap and the segmentation areas of the multiple windows all contain pixel point i, the above-mentioned summation is performed on the pixel of the multiple windows and the average is taken.

[0123] In one embodiment of the present invention, the sharpening module 30 first performs pixel-by-pixel subtraction between the original endoscopic image and the base layer image to obtain a detail layer image, and the calculation formula is as follows:

[0124] D i =I i -Q i

[0125] Where D represents the detail layer image.

[0126] Then, the sharpening module 30 weightedly synthesizes the original endoscopic image and the detail layer image by the following formula to obtain a detail enhanced image:

[0127] R i =I i +η k ·D i

[0128] Where R represents the detail enhanced image, η k Indicates the sharpening factor of the kth window segmentation area where the i-th pixel is located. The sharpening factor corresponding to each window is the same, and all pixels in each window segmentation area are calculated using the corresponding sharpening factor.

[0129] Furthermore, the real-time endoscopic image detail enhancement device of the embodiment of the present invention may further include an updating module, which is used to adaptively update the edge adjustment factor, the detail adjustment factor and the sharpening factor through an adaptive parameter updating algorithm.

[0130] Specifically, the update module updates the edge adjustment factor by the following formula:

[0131] λ=(0.001×L) 2

[0132] Among them, L is the dynamic range of each window segmentation area.

[0133] The update module updates the detail adjustment factor by the following formula:

[0134]

[0135] Among them, x is 0.3 times the mean value of the pixels in each window segmentation area, and m is calculated by the following formula:

[0136]

[0137] The update module updates the sharpening factor by the following formula:

[0138] η k =γ k +4.5.

[0139] In a specific embodiment of the present invention, the original endoscopic image is shown in FIG2(a), which is an RGB image with a resolution of 548*548. The guided filtering uses a circular window set with a radius of 8 pixels. After being processed by the filtering module 20, a base layer image as shown in FIG2(b) is obtained. Then, after being processed by the sharpening module 30, a detail-enhanced image as shown in FIG2(c) is obtained. It can be seen that after the detail enhancement is performed using the real-time endoscopic image detail enhancement device of the embodiment of the present invention, the subtle texture and edge features in the endoscopic image are significantly enhanced, the blood vessel texture becomes very clear, and even the tiny details inside the blood vessel can be seen, and the blood vessel color becomes more vivid; at the same time, the bright spot edge in the image presents a clearer and sharper outline, and the bright spot is fuller and brighter, which greatly improves the quality of the endoscopic image.

[0140] According to the real-time endoscopic image detail enhancement device of the embodiment of the present invention, the original endoscopic image is guided filtered by a detail enhancement guided filtering processing algorithm to obtain a base layer image, and the original endoscopic image and the base layer image are mask sharpening processed by an unsharp mask processing algorithm to obtain a detail enhanced image, which has low computational complexity and good real-time performance, and can highlight the details and edges of the image and retain the overall brightness and color information of the image, thereby being able to quickly and effectively improve the quality of the endoscopic image.

[0141] The present invention and its embodiments are described schematically above, and the description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. Therefore, if a person skilled in the art is inspired by it and designs a structural method and an embodiment similar to the technical solution without creativity without departing from the purpose of the invention, they shall all fall within the protection scope of the present invention.

Claims

1. A real-time endoscopic image detail enhancement method, characterized in that: The following steps are involved: S1, obtain the original endoscopic image; S2, performing guided filtering on the original endoscopic image by a detail enhancement guided filtering processing algorithm to obtain a base layer image; S3, performing mask sharpening processing on the original endoscopic image and the base layer image by using an unsharp mask processing algorithm to obtain a detail enhanced image.

2. The real-time endoscopic image detail enhancement method according to claim 1, characterized in that: Step S2 specifically includes: S21, segmenting the original endoscopic image through a two-dimensional window set; S22, calculating the mean and variance of the pixels in each window segmentation area; S23, obtaining edge features corresponding to each window by the following formula, and calculating edge perception factors of each window according to the edge features: in, represents the edge feature corresponding to the kth window, represents the local variance of the k-th window, G(·) represents the Gaussian filter, |grad| represents the absolute value of the gradient, Γ(k) represents the edge perception factor of the k-th window, N is the total number of pixels of the original endoscopic image, and λ is the edge adjustment factor; S24, according to the mean and variance of the pixels in each window segmentation area and the edge perception factor of each window, the guidance weight corresponding to each window is calculated by the following formula: Among them, a k and b k is the guidance weight corresponding to the kth window, σ 2 k is the variance of the pixels in the kth window segmentation area, ε is the regularization factor, γ k is the detail adjustment factor, u k is the mean value of the pixels in the k-th window segmentation area; S25, according to the guidance weight corresponding to each window, the base layer image is calculated by the following formula: Among them, i is the serial number of the pixel point, k:i∈ωk represents the kth window segmentation area containing the pixel point i, |ω| represents the total number of windows containing the pixel i, I represents the original endoscopic image, and Q represents the base layer image.

3. The real-time endoscopic image detail enhancement method according to claim 2, characterized in that: Step S3 specifically includes: S31, performing pixel-by-pixel subtraction between the original endoscopic image and the base layer image to obtain a detail layer image; S32, weighted synthesis of the original endoscopic image and the detail layer image by the following formula to obtain a detail enhanced image: R i =I i +n k ·D i Wherein, R represents the detail enhanced image, D represents the detail layer image, and ηk represents the sharpening factor of the kth window segmentation area where the ith pixel is located.

4. The real-time endoscopic image detail enhancement method according to claim 3, characterized in that: Also includes: The edge adjustment factor, the detail adjustment factor and the sharpening factor are adaptively updated through an adaptive parameter updating algorithm.

5. The real-time endoscopic image detail enhancement method according to claim 4, characterized in that: The edge adjustment factor is updated by the following formula: λ=(0.001×L) 2 Among them, L is the dynamic range of each window segmentation area; The detail adjustment factor is updated by the following formula: Among them, x is 0.3 times the mean value of the pixels in each window segmentation area, and m is calculated by the following formula: The sharpening factor is updated by: or k =c k +4.

5.

6. A real-time endoscopic image detail enhancement device, characterized in that: include: An acquisition module, wherein the acquisition module is used to acquire an original endoscopic image; A filtering module, the filtering module is used to perform guided filtering on the original endoscopic image through a detail enhancement guided filtering processing algorithm to obtain a base layer image; A sharpening module is used to perform mask sharpening processing on the original endoscopic image and the base layer image through an unsharp mask processing algorithm to obtain a detail-enhanced image.

7. The real-time endoscopic image detail enhancement device according to claim 6, characterized in that: The filtering module is specifically used for: Segmenting the original endoscopic image through a two-dimensional window set; Calculate the mean and variance of the pixels in each window segmentation area; The edge features corresponding to each window are obtained by the following formula, and the edge perception factor of each window is calculated according to the edge features: in, represents the edge feature corresponding to the kth window, represents the local variance of the k-th window, G(·) represents the Gaussian filter, |grad| represents the absolute value of the gradient, Γ(k) represents the edge perception factor of the k-th window, N is the total number of pixels of the original endoscopic image, and λ is the edge adjustment factor; According to the mean and variance of the pixels in each window segmentation area and the edge perception factor of each window, the guidance weight corresponding to each window is calculated by the following formula: Among them, a k and b k is the guidance weight corresponding to the kth window, σ 2 k is the variance of the pixels in the kth window segmentation area, ε is the regularization factor, γ k is the detail adjustment factor, u k is the mean value of the pixels in the k-th window segmentation area; According to the guidance weight corresponding to each window, the base layer image is calculated by the following formula: Among them, i is the serial number of the pixel point, k:i∈ω k represents the kth window segmentation area including pixel point i, |ω| represents the total number of windows including pixel i, I represents the original endoscopic image, and Q represents the base layer image.

8. The real-time endoscopic image detail enhancement device according to claim 7, characterized in that: The sharpening module is specifically used for: Subtracting the original endoscopic image from the base layer image pixel by pixel to obtain a detail layer image; The original endoscopic image and the detail layer image are weighted and synthesized by the following formula to obtain a detail enhanced image: R i =I i +n k ·D i Wherein, R represents the detail enhanced image, D represents the detail layer image, and n k Indicates the sharpening factor of the k-th window segmentation area where the i-th pixel is located.

9. The real-time endoscopic image detail enhancement device according to claim 8, characterized in that: Also includes: An updating module, wherein the updating module is used to adaptively update the edge adjustment factor, the detail adjustment factor and the sharpening factor through an adaptive parameter updating algorithm.

10. The real-time endoscopic image detail enhancement device according to claim 9, characterized in that: The updating module updates the edge adjustment factor by the following formula: λ=(0.001×L) 2 Among them, L is the dynamic range of each window segmentation area; The updating module updates the detail adjustment factor by the following formula: Among them, x is 0.3 times the mean value of the pixels in each window segmentation area, and m is calculated by the following formula: The updating module updates the sharpening factor by the following formula: or k =c k +4.5.

Images