An endoscope-based assisted detection method and system

Abstract

The application discloses an endoscope-based auxiliary detection method and system, and relates to the technical field of image enhancement. The endoscope image is decomposed to obtain a basic layer image and a detail layer image; the basic layer image is subjected to brightness enhancement to obtain a first enhancement component; the detail layer image is subjected to edge strengthening to obtain a second enhancement component; the first enhancement component and the second enhancement component are integrated to obtain a first target image; the first target image is subjected to contrast enhancement to obtain a second target image; and the second target image is displayed on a target terminal. By separating the color distribution information and the structure information of the image, the brightness is subjected to gamma correction, the structure details are subjected to edge enhancement, and then fusion processing is performed, and the overall contrast is improved in combination with histogram equalization, so that the problems of uneven illumination and fuzzy details of the endoscope image are solved, and the lesion visibility is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of image enhancement technology, and more specifically to an endoscope-based auxiliary detection method and system. Background Technology

[0002] With the advancement of medical technology, endoscopy has been widely used in clinical diagnosis. Endoscopes are extensively applied in the diagnosis and treatment of various diseases affecting the gastrointestinal, respiratory, and urinary systems. By inserting a thin, tubular device into the human body, a camera captures real-time images, which are then transmitted to an external monitor. This allows doctors to visually observe internal tissue structures, enabling early detection of lesions and the development of treatment plans.

[0003] However, while endoscopic technology can provide real-time images and effectively observe lesion areas, its image quality is often limited by the complex tissue structure within the human body. In practical applications, due to factors such as uneven distribution of human tissues and differences in surface reflectivity, it is difficult to maintain consistent light intensity and uniformity in different areas under endoscopy, leading to problems such as overexposure and shadows in endoscopic images. Poor image quality results in weak visualization of lesions, and insufficient clarity in displaying fine internal tissues and mild lesions. Summary of the Invention

[0004] The purpose of this invention is to solve the problem of poor lesion visibility mentioned in the background art, and to propose an endoscope-based auxiliary detection method and system.

[0005] A first aspect of this invention provides an endoscope-based assisted detection method, the method comprising:

[0006] Acquire raw images captured by the endoscope;

[0007] The original image is decomposed using guided filtering to obtain a base layer image and a detail layer image;

[0008] The base layer image is converted to HSV space, the V component is gamma corrected, and then converted back to RGB image to obtain the first enhanced component.

[0009] Edge enhancement is performed on the detail layer image to obtain the second enhancement component;

[0010] The first enhancement component and the second enhancement component are integrated to obtain the first target image;

[0011] The first target image is contrast-enhanced to obtain the second target image;

[0012] The second target image is displayed on the target terminal.

[0013] Optionally, gamma correction of the V component includes:

[0014] The first luminance frequency distribution function is obtained based on the luminance value of each pixel in the V component.

[0015] Based on the maximum and minimum brightness values ​​of the V component, multiple brightness ranges are determined, namely, low brightness range, medium brightness range and high brightness range;

[0016] Based on the distribution function of the first brightness frequency, the average frequency of each brightness interval is calculated and used as the coverage threshold of each brightness interval.

[0017] Based on the coverage threshold of each brightness range, the first brightness frequency distribution function is modified to obtain the second brightness frequency distribution function:

[0018]

[0019] Where F1 and F2 are the first and second luminance frequency distribution functions, respectively; b is the luminance value; BR1, BR2 and BR3 are luminance intervals; CT1, CT2 and CT3 are the coverage thresholds for each luminance interval.

[0020] Calculate the cumulative probability distribution function based on the second brightness frequency distribution function;

[0021] Calculate the gamma parameter corresponding to each brightness value based on the cumulative probability distribution function;

[0022] Based on the gamma parameters corresponding to each brightness value, the V component is gamma-corrected to obtain an enhanced V component.

[0023] Optionally, calculating the gamma parameter corresponding to each brightness value based on the cumulative probability distribution function includes:

[0024] γ(b) = max{1-CDF(b), α}

[0025] Where γ(b) is the gamma parameter value corresponding to brightness b; CDF is the cumulative probability distribution function; α is the preset lower threshold; and max indicates taking the maximum value.

[0026] Optionally, the edge enhancement of the detail layer image to obtain the second enhancement component includes:

[0027] Gaussian blur is applied to the detail layer image to obtain the blur components;

[0028] Subtract the blur component from the detail layer image to obtain the detail component;

[0029] The detail component is weighted and summed with the detail layer image, and then thresholded to obtain the second enhancement component.

[0030] Optionally, the step of enhancing the contrast of the first target image to obtain the second target image includes:

[0031] The first target image is converted to Lab space, and discrete wavelet transform is performed on the L component to obtain the low-frequency component image and the high-frequency component image.

[0032] Semi-soft thresholding denoising is performed on the high-frequency component image to obtain an enhanced high-frequency component image; adaptive histogram equalization is performed on the low-frequency component image to obtain an enhanced low-frequency component image.

[0033] Perform inverse wavelet transform on the enhanced high-frequency component image and the low-frequency component image to obtain the enhanced L component; convert back to RGB image to obtain the second target image.

[0034] A second aspect of this invention provides an endoscope-based auxiliary detection system, the system comprising:

[0035] The data acquisition module is used to acquire raw images captured by the endoscope;

[0036] The image decomposition module is used to decompose the original image using guided filtering to obtain a base layer image and a detail layer image;

[0037] The brightness enhancement module is used to convert the base layer image to HSV space, perform gamma correction on the V component, and then convert it back to RGB image to obtain the first enhancement component.

[0038] The edge enhancement module is used to enhance the edges of the detail layer image to obtain the second enhancement component;

[0039] An image reconstruction module is used to integrate the first enhancement component and the second enhancement component to obtain a first target image;

[0040] A contrast enhancement module is used to enhance the contrast of the first target image to obtain a second target image;

[0041] A visualization module is used to display the second target image on the target terminal.

[0042] Optionally, the brightness enhancement module includes:

[0043] The frequency statistics module is used to obtain the first luminance frequency distribution function based on the luminance value of each pixel in the V component.

[0044] The brightness segmentation module is used to determine multiple brightness ranges based on the maximum and minimum brightness values ​​of the V component, namely, the low brightness range, the medium brightness range, and the high brightness range.

[0045] The threshold calculation module is used to calculate the average frequency of each brightness interval based on the distribution function of the first brightness frequency, and use it as the coverage threshold of each brightness interval.

[0046] The distribution adjustment module is used to modify the first brightness frequency distribution function according to the coverage threshold of each brightness interval to obtain the second brightness frequency distribution function:

[0047]

[0048] Where F1 and F2 are the first and second luminance frequency distribution functions, respectively; b is the luminance value; BR1, BR2 and BR3 are luminance intervals; CT1, CT2 and CT3 are the coverage thresholds for each luminance interval.

[0049] The cumulative probability calculation module is used to calculate the cumulative probability distribution function based on the second brightness frequency distribution function;

[0050] The gamma parameter calculation module is used to calculate the gamma parameter corresponding to each brightness value based on the cumulative probability distribution function.

[0051] An adaptive adjustment module is used to perform gamma correction on the V component based on the gamma parameters corresponding to each brightness value, so as to obtain an enhanced V component.

[0052] Optionally, the calculation process of the gamma parameter calculation module includes:

[0053] γ(b) = max{1-CDF(b), α}

[0054] Where γ(b) is the gamma parameter value corresponding to brightness b; CDF is the cumulative probability distribution function; α is the preset lower threshold; and max indicates taking the maximum value.

[0055] Optionally, the edge enhancement module includes:

[0056] A Gaussian smoothing module is used to apply Gaussian blur to the detail layer image to obtain a blur component;

[0057] An edge extraction module is used to subtract the blur component from the detail layer image to obtain the detail component;

[0058] The detail enhancement module is used to perform a weighted summation of the detail component and the detail layer image, and then perform threshold cropping to obtain the second enhancement component.

[0059] Optionally, the contrast enhancement module includes:

[0060] The first spatial conversion module is used to convert the first target image to Lab space;

[0061] The wavelet transform module is used to perform discrete wavelet transform on the L component to obtain low-frequency component images and high-frequency component images.

[0062] The high-frequency enhancement module is used to perform semi-soft thresholding denoising on the high-frequency component image to obtain an enhanced high-frequency component image.

[0063] The low-frequency enhancement module is used to perform adaptive histogram equalization on the low-frequency component image to obtain an enhanced low-frequency component image.

[0064] The component recovery module is used to perform inverse wavelet transform on the enhanced high-frequency component image and the low-frequency component image to obtain the enhanced L component.

[0065] The second spatial conversion module is used to convert the a component, b component and enhanced L component into RGB images to obtain the second target image.

[0066] The beneficial effects of this invention are:

[0067] This invention proposes an endoscope-based assisted detection method, which includes: acquiring an original image captured by an endoscope; decomposing the original image using guided filtering to obtain a base layer image and a detail layer image; converting the base layer image to HSV space, performing gamma correction on the V component, and then converting it back to an RGB image to obtain a first enhancement component; performing edge enhancement on the detail layer image to obtain a second enhancement component; integrating the first enhancement component and the second enhancement component to obtain a first target image; performing contrast enhancement on the first target image to obtain a second target image; and displaying the second target image on a target terminal.

[0068] By separating the color distribution information and structural information of the image, gamma correction is performed on the brightness and edge enhancement is performed on the structural details, and then the images are fused together. Combined with histogram equalization, the overall contrast is improved, which solves problems such as uneven illumination and blurred details in endoscopic images, thereby significantly improving the visibility of lesions. Attached Figure Description

[0069] Figure 1 A flowchart of an endoscope-based assisted detection method is provided for an embodiment of the present invention;

[0070] Figure 2 This invention provides an architecture diagram of an endoscope-based auxiliary detection system. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the 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.

[0072] This invention provides an endoscope-based assisted detection method. See also... Figure 1 , Figure 1 A flowchart illustrating an endoscope-based assisted detection method provided in an embodiment of the present invention. The method includes the following steps:

[0073] S101, acquire the raw images captured by the endoscope;

[0074] S102, guided filtering is used to decompose the original image to obtain the base layer image and detail layer image.

[0075] S103, convert the base layer image to HSV space, perform gamma correction on the V component, and then convert it back to RGB image to obtain the first enhancement component.

[0076] S104, edge enhancement is performed on the detail layer image to obtain the second enhancement component.

[0077] S105, the first enhancement component and the second enhancement component are integrated to obtain the first target image.

[0078] S106, perform contrast enhancement on the first target image to obtain the second target image.

[0079] S107, the second target image is displayed on the target terminal.

[0080] The endoscopic-based auxiliary detection method provided by this invention separates the color distribution information and structural information of the image, performs gamma correction on the brightness and edge enhancement on the structural details, and then performs fusion processing. Combined with histogram equalization, the overall contrast is improved, which solves the problems of uneven illumination and blurred details in endoscopic images, thereby significantly improving the visibility of lesions.

[0081] In one embodiment, step S102 includes:

[0082] Step 1: Filter the original image using a guided filter with a window size of 5×5 to obtain the base layer image. Step 2: Subtract the base layer image from the original image to obtain the detail layer image. By separating the base layer and the detail layer, different characteristics of the image can be processed separately, enhancing the quality of the final image.

[0083] In one embodiment, step S103 includes:

[0084] Step 1: Based on the luminance value of each pixel in the V component, obtain the first luminance frequency distribution function F1. F1(b) = C indicates that the luminance value b appears C times in the V component.

[0085] Step two: Based on the maximum brightness value BMAX and minimum brightness value BMIN of the V component, determine multiple brightness ranges, namely the low brightness range, medium brightness range, and high brightness range. Specifically, the entire brightness range can be equally divided into three intervals based on the maximum and minimum brightness values, in ascending order: low brightness range BR1, medium brightness range BR2, and high brightness range BR3.

[0086] Step 3: Based on the distribution function of the first brightness frequency, calculate the average frequency of each brightness interval, which serves as the coverage threshold for each brightness interval. Specifically, for any brightness interval, denoted as the target brightness interval, the formula for calculating the coverage threshold CT is: Where start and end are the minimum and maximum values ​​of the target brightness range, respectively.

[0087] Step 4: Based on the coverage threshold of each brightness range, modify the first brightness frequency distribution function to obtain the second brightness frequency distribution function:

[0088]

[0089] Where F1 and F2 are the first and second luminance frequency distribution functions, respectively; b is the luminance value; BR1, BR2 and BR3 are luminance intervals; CT1, CT2 and CT3 are the coverage thresholds of each luminance interval.

[0090] Step 5: Calculate the cumulative probability distribution function based on the second brightness frequency distribution function. Specifically, first calculate the probability distribution for each brightness level. Then calculate the cumulative probability distribution.

[0091]

[0092] Step 6: Calculate the gamma parameter corresponding to each brightness value based on the cumulative probability distribution function. Specifically, γ(b) = max{1-CDF(b), α}. Where γ(b) is the gamma parameter value corresponding to brightness b; CDF is the cumulative probability distribution function; α is a preset lower threshold; and max indicates taking the maximum value.

[0093] Step seven: Based on the gamma parameters corresponding to each brightness value, perform gamma correction on the V component to obtain the enhanced V component, denoted as EV. Specifically, Where (x,y) are the position coordinates; b is the brightness value at position (x,y).

[0094] In one implementation, gamma correction is performed based on the frequency distribution of the luminance values ​​of the V component. This allows for adaptive adjustment of the luminance enhancement process according to the luminance distribution of the image itself, rather than using a uniform parameter. This adaptive approach can avoid over-enhancement or under-enhancement based on the specific luminance characteristics of the image, thereby improving the overall image quality.

[0095] In one implementation, threshold pruning of the frequency distribution avoids excessively drastic changes in the cumulative probability distribution, resulting in more stable changes in the gamma parameter and thus achieving better visual effects.

[0096] In one implementation, the lower threshold can be set to 0.5. By setting a lower threshold, over-enhancing the image can be avoided, which could lead to loss of detail.

[0097] In one embodiment, step S104 includes:

[0098] Step 1: Apply Gaussian blur to the detail layer image to obtain the blur component. Specifically, a Gaussian kernel with a window size of 5×5 and a standard deviation of 1 can be used.

[0099] Step 2: Subtract the blur component from the detail layer image to obtain the detail component.

[0100] Step 3: Weight the detail component and the detail layer image, then perform a threshold cropping to obtain the second enhancement component. Specifically, the weights of both can be set to 0.8. After summing, values ​​less than 0 are changed to 0, and values ​​greater than 255 are changed to 255.

[0101] In one embodiment, in step S105, the first enhancement component and the second enhancement component are integrated to obtain the first target image, which can be achieved by weighted summation. The weight coefficient of the first enhancement component is set to 1; the weight coefficient of the second enhancement component is between 0.2 and 0.5, and can be flexibly selected. Through appropriate weighted summation, the final first target image can achieve a balance between brightness and detail, making the image clearer, more delicate, rich in detail, and realistic.

[0102] In one embodiment, step S106 includes:

[0103] Step 1: Convert the first target image to Lab space to obtain the L component, a component and b component.

[0104] Step 2: Perform discrete wavelet transform on the L component to obtain the low-frequency component image and the high-frequency component image.

[0105] Step 3: Perform semi-soft thresholding denoising on the high-frequency component image to obtain an enhanced high-frequency component image.

[0106] Step four: Perform adaptive histogram equalization on the low-frequency component image to obtain an enhanced low-frequency component image.

[0107] Step 5: Perform inverse wavelet transform on the enhanced high-frequency component image and the low-frequency component image to obtain the enhanced L component.

[0108] Step six: Convert the a component, b component, and enhanced L component into RGB images to obtain the second target image.

[0109] This embodiment optimizes the L-component of the image through multiple steps, including wavelet transform, denoising, and histogram equalization, resulting in a significant enhancement in image contrast. This leads to a more balanced overall image perception, with good brightness and contrast across the entire image, while also improving detail areas, thereby enhancing image visibility.

[0110] This invention provides an endoscope-based auxiliary detection system. See also... Figure 2 , Figure 2 This is an architectural diagram of an endoscope-based auxiliary detection system provided in an embodiment of the present invention. The system includes:

[0111] The data acquisition module is used to acquire raw images captured by the endoscope.

[0112] The image decomposition module is used to decompose the original image using guided filtering to obtain the base layer image and the detail layer image.

[0113] The brightness enhancement module is used to convert the base layer image to HSV space, perform gamma correction on the V component, and then convert it back to an RGB image to obtain the first enhanced component.

[0114] The edge enhancement module is used to enhance the edges of the detail layer image to obtain the second enhancement component.

[0115] The image reconstruction module is used to integrate the first enhancement component and the second enhancement component to obtain the first target image.

[0116] The contrast enhancement module is used to enhance the contrast of the first target image to obtain the second target image.

[0117] The visualization module is used to display the second target image on the target terminal.

[0118] The endoscopic-based auxiliary detection method provided by this invention separates the color distribution information and structural information of the image, performs gamma correction on the brightness and edge enhancement on the structural details, and then performs fusion processing. Combined with histogram equalization, the overall contrast is improved, which solves the problems of uneven illumination and blurred details in endoscopic images, thereby significantly improving the visibility of lesions.

[0119] In one embodiment, the brightness enhancement module includes:

[0120] The frequency statistics module is used to obtain the first luminance frequency distribution function based on the luminance value of each pixel in the V component.

[0121] The brightness segmentation module is used to determine multiple brightness ranges based on the maximum and minimum brightness values ​​of the V component, namely, the low brightness range, the medium brightness range, and the high brightness range.

[0122] The threshold calculation module is used to calculate the average frequency of each brightness interval based on the distribution function of the first brightness frequency, and use it as the coverage threshold of each brightness interval.

[0123] The distribution adjustment module is used to modify the first brightness frequency distribution function according to the coverage threshold of each brightness interval to obtain the second brightness frequency distribution function.

[0124]

[0125] Where F1 and F2 are the first and second luminance frequency distribution functions, respectively; b is the luminance value; BR1, BR2 and BR3 are luminance intervals; CT1, CT2 and CT3 are the coverage thresholds of each luminance interval.

[0126] The cumulative probability calculation module is used to calculate the cumulative probability distribution function based on the second brightness frequency distribution function.

[0127] The gamma parameter calculation module is used to calculate the gamma parameters corresponding to each brightness value based on the cumulative probability distribution function.

[0128] The adaptive adjustment module is used to perform gamma correction on the V component based on the gamma parameters corresponding to each brightness value, so as to obtain an enhanced V component.

[0129] In one embodiment, the edge enhancement module includes:

[0130] The Gaussian smoothing module is used to apply Gaussian blur to the detail layer image to obtain the blur component.

[0131] The edge extraction module is used to subtract the blur component from the detail layer image to obtain the detail component.

[0132] The detail enhancement module is used to perform a weighted summation of the detail components and the detail layer image, and then perform threshold cropping to obtain the second enhancement component.

[0133] In one embodiment, the contrast enhancement module includes:

[0134] The first spatial conversion module is used to convert the first target image to Lab space.

[0135] The wavelet transform module is used to perform discrete wavelet transform on the L component to obtain low-frequency component images and high-frequency component images.

[0136] The high-frequency enhancement module is used to perform semi-soft thresholding denoising on the high-frequency component image to obtain an enhanced high-frequency component image.

[0137] The low-frequency enhancement module is used to perform adaptive histogram equalization on the low-frequency component image to obtain an enhanced low-frequency component image.

[0138] The component recovery module is used to perform inverse wavelet transform on the enhanced high-frequency component image and the low-frequency component image to obtain the enhanced L component.

[0139] The second spatial conversion module is used to convert the a component, b component and enhanced L component into RGB images to obtain the second target image.

[0140] It should be noted that, in this document, terms such as “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0141] The embodiments of the present invention have been described in detail above, but the content described is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. An endoscope-based assisted detection method, characterized in that, The method includes: Acquire raw images captured by the endoscope; The original image is decomposed using guided filtering to obtain a base layer image and a detail layer image; The base layer image is converted to HSV space, the V component is gamma corrected, and then converted back to RGB image to obtain the first enhanced component. Edge enhancement is performed on the detail layer image to obtain the second enhancement component; The first enhancement component and the second enhancement component are integrated to obtain the first target image; The first target image is contrast-enhanced to obtain the second target image; Display the second target image on the target terminal; Among them, gamma correction of the V component includes: The first luminance frequency distribution function is obtained based on the luminance value of each pixel in the V component. Based on the maximum and minimum brightness values ​​of the V component, multiple brightness ranges are determined, namely, low brightness range, medium brightness range and high brightness range; Based on the distribution function of the first brightness frequency, the average frequency of each brightness interval is calculated and used as the coverage threshold of each brightness interval. Based on the coverage threshold of each brightness range, the first brightness frequency distribution function is modified to obtain the second brightness frequency distribution function: ； Where F1 and F2 are the first and second luminance frequency distribution functions, respectively; b is the luminance value; BR1, BR2 and BR3 are luminance intervals; CT1, CT2 and CT3 are the coverage thresholds for each luminance interval. Calculate the cumulative probability distribution function based on the second brightness frequency distribution function; Calculate the gamma parameter corresponding to each brightness value based on the cumulative probability distribution function; Based on the gamma parameters corresponding to each brightness value, the V component is gamma-corrected to obtain an enhanced V component.

2. The endoscope-based assisted detection method according to claim 1, characterized in that, The step of calculating the gamma parameters corresponding to each brightness value based on the cumulative probability distribution function includes: ； Where γ(b) is the gamma parameter value corresponding to brightness b; CDF is the cumulative probability distribution function; It is the preset lower threshold; max indicates the maximum value.

3. The endoscope-based assisted detection method according to claim 1, characterized in that, The edge enhancement of the detail layer image to obtain the second enhancement component includes: Gaussian blur is applied to the detail layer image to obtain the blur components; Subtract the blur component from the detail layer image to obtain the detail component; The detail component is weighted and summed with the detail layer image, and then thresholded to obtain the second enhancement component.

4. The endoscope-based assisted detection method according to claim 1, characterized in that, The step of enhancing the contrast of the first target image to obtain the second target image includes: The first target image is converted to Lab space, and discrete wavelet transform is performed on the L component to obtain the low-frequency component image and the high-frequency component image. Semi-soft thresholding denoising is performed on the high-frequency component image to obtain an enhanced high-frequency component image; adaptive histogram equalization is performed on the low-frequency component image to obtain an enhanced low-frequency component image. Perform inverse wavelet transform on the enhanced high-frequency component image and the low-frequency component image to obtain the enhanced L component; convert back to RGB image to obtain the second target image.

5. An endoscope-based auxiliary detection system, characterized in that, The system includes: The data acquisition module is used to acquire raw images captured by the endoscope; The image decomposition module is used to decompose the original image using guided filtering to obtain a base layer image and a detail layer image; The brightness enhancement module is used to convert the base layer image to HSV space, perform gamma correction on the V component, and then convert it back to RGB image to obtain the first enhancement component. The edge enhancement module is used to enhance the edges of the detail layer image to obtain the second enhancement component; An image reconstruction module is used to integrate the first enhancement component and the second enhancement component to obtain a first target image; A contrast enhancement module is used to enhance the contrast of the first target image to obtain a second target image; A visualization module is used to display the second target image on the target terminal; The brightness enhancement module includes: The frequency statistics module is used to obtain the first luminance frequency distribution function based on the luminance value of each pixel in the V component. The brightness segmentation module is used to determine multiple brightness ranges based on the maximum and minimum brightness values ​​of the V component, namely, the low brightness range, the medium brightness range, and the high brightness range. The threshold calculation module is used to calculate the average frequency of each brightness interval based on the distribution function of the first brightness frequency, and use it as the coverage threshold of each brightness interval. The distribution adjustment module is used to modify the first brightness frequency distribution function according to the coverage threshold of each brightness interval to obtain the second brightness frequency distribution function: ； Where F1 and F2 are the first and second luminance frequency distribution functions, respectively; b is the luminance value; BR1, BR2 and BR3 are luminance intervals; CT1, CT2 and CT3 are the coverage thresholds for each luminance interval. The cumulative probability calculation module is used to calculate the cumulative probability distribution function based on the second brightness frequency distribution function; The gamma parameter calculation module is used to calculate the gamma parameter corresponding to each brightness value based on the cumulative probability distribution function. An adaptive adjustment module is used to perform gamma correction on the V component based on the gamma parameters corresponding to each brightness value, so as to obtain an enhanced V component.

6. The endoscope-based auxiliary detection system according to claim 5, characterized in that, The calculation process of the gamma parameter calculation module includes: ； Where γ(b) is the gamma parameter value corresponding to brightness b; CDF is the cumulative probability distribution function; It is the preset lower threshold; max indicates the maximum value.

7. The endoscope-based auxiliary detection system according to claim 5, characterized in that, The edge enhancement module includes: A Gaussian smoothing module is used to apply Gaussian blur to the detail layer image to obtain a blur component; An edge extraction module is used to subtract the blur component from the detail layer image to obtain the detail component; The detail enhancement module is used to perform a weighted summation of the detail component and the detail layer image, and then perform threshold cropping to obtain the second enhancement component.

8. The endoscope-based auxiliary detection system according to claim 5, characterized in that, The contrast enhancement module includes: The first spatial conversion module is used to convert the first target image to Lab space; The wavelet transform module is used to perform discrete wavelet transform on the L component to obtain low-frequency component images and high-frequency component images. The high-frequency enhancement module is used to perform semi-soft thresholding denoising on the high-frequency component image to obtain an enhanced high-frequency component image. The low-frequency enhancement module is used to perform adaptive histogram equalization on the low-frequency component image to obtain an enhanced low-frequency component image. The component recovery module is used to perform inverse wavelet transform on the enhanced high-frequency component image and the low-frequency component image to obtain the enhanced L component. The second spatial conversion module is used to convert the a component, b component and enhanced L component into RGB images to obtain the second target image.

Images