Endoscope display enhancement method and system

Abstract

The invention belongs to the technical field of image recognition, and provides an endoscope display enhancement method and system, and the method comprises the steps: obtaining an original image of an endoscope, and carrying out the processing of the original image according to a preset algorithm, and obtaining an enhanced image; extracting a target organ image based on the enhanced image; detecting a lesion area in the target organ image according to a preset recognition algorithm, and evaluating the lesion condition of the target organ according to the lesion area; the lesion condition is combined with the target organ image as the auxiliary diagnosis information, and visual display is performed in combination with the preset interface layout principle, so that the visual effect of the endoscope image can be effectively improved, and reliable auxiliary diagnosis information can be provided for medical staff.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to an endoscope display enhancement method and system. Background Art

[0002] In the medical field, endoscopes are usually used to examine organ lesions. During the endoscopic imaging process, the relative movement between the camera and the examined organ, electronic noise, interference and other factors result in poor visual effects of the endoscopic image.

[0003] At the same time, existing organ diseases are highly complex. Doctors are required to have strong professional knowledge and combine a variety of examination information to accurately diagnose patients. This means that when doctors use endoscopes to scan organ images, they need to concentrate on observing abnormal areas in the image and also look at medical images provided by other examination equipment when abnormal areas are found.

[0004] Obviously, the above method is not only cumbersome to operate and easy for doctors to get tired, but also due to the inconsistent professional and technical levels of doctors, it is likely to cause inconsistent diagnosis results, that is, the reliability of the diagnosis results is difficult to guarantee. Summary of the invention

[0005] In view of this, an embodiment of the present application provides an endoscope display enhancement method and system to solve the above-mentioned technical problems.

[0006] A first aspect of an embodiment of the present application provides an endoscope display enhancement method, comprising:

[0007] Acquire the original image of the endoscope, and process the original image according to a preset algorithm to obtain an enhanced image;

[0008] Extracting the target organ image based on the enhanced image;

[0009] According to the preset recognition algorithm, the lesion area in the target organ image is detected, and the lesion condition of the target organ is evaluated based on the lesion area;

[0010] The lesion condition is combined with the target organ image as auxiliary diagnostic information and visualized according to the preset interface layout principles.

[0011] For example, the step of obtaining an original endoscope image and processing the original image according to a preset algorithm to obtain an enhanced image includes:

[0012] Stretching the corresponding channels of the endoscope raw image in the first color space;

[0013] Converting the stretched first color space image to a second color space, and stretching the corresponding channels in the second color space;

[0014] The stretched second color space image is converted to the first color space to obtain an enhanced image.

[0015] For example, the endoscope original image is stretched for the corresponding channel in the first color space, wherein the first color space is the RGB color space, and the mapping function for contrast stretching the green channel in the RGB color space is expressed by the following formula:

[0016] g * (x,y)=[1+d g ×((g(x,y)) -1 -1) 2 ] -1

[0017] In the above formula, g(x, y) and g * (x, y) are the normalized grayscale values ​​of the green channel of the original image and the enhanced image at the pixel point (x, y); d g It is the green channel grayscale mapping parameter.

[0018] For example, the stretched first color space image is converted to a second color space, and the corresponding channel is stretched in the second color space, wherein the second color space is an HSV color space, and the mapping function for stretching the grayscale contrast of the saturation channel in the second color space is expressed by the following formula:

[0019] s * =[1+d s ×(s -1 -1) 2 ] -1

[0020] In the above formula, s and s * are the grayscale values ​​of the saturation channel before and after mapping; d s It is the S channel grayscale mapping parameter.

[0021] For example, the step of extracting a target organ image based on an enhanced image includes:

[0022] The enhanced image is processed into superpixels by a preset segmentation algorithm to obtain a number of superpixels;

[0023] Superpixels are classified based on convolutional neural networks to distinguish between superpixels inside organs and superpixels at the edge of organs.

[0024] Select several seed points of organ edge superpixels to construct the initial contour;

[0025] Based on the initial contour, the target organ image is segmented and extracted by solving the minimum value of the preset function.

[0026] For example, the superpixels are classified based on the convolutional neural network to distinguish between superpixels inside the organ and superpixels at the edge of the organ, wherein the objective function of the convolutional neural network is:

[0027]

[0028] In the above formula, represents the probability distribution of the pixel x output by the soft-max layer of the convolutional neural network, i represents the number of layers of the convolutional neural network, and λ i is the weight factor of the i-th layer, N is the total number of pixels, N i is a category L n , k represents the basic fact category label, and k∈[0,1] represents the basic fact category label, where 0 represents the superpixel inside the organ, 1 represents the superpixel at the edge of the organ, and x is the pixel point.

[0029] For example, the method of detecting the lesion area in the target organ image according to the preset recognition algorithm and evaluating the lesion condition of the target organ according to the lesion area is specifically as follows:

[0030] Acquire a number of lesion images, and extract the depth features of the lesion images one by one;

[0031] The deep features of each lesion image are encoded into multiple groups of integers through a hash algorithm and stored in a Bloom filter;

[0032] When receiving the image of the target organ to be detected, extracting the depth feature of the image of the organ to be detected;

[0033] The deep features of the image of the organ to be detected are encoded by a local sensitive hash function, and then a proximity search is performed in a Bloom filter to evaluate the pathological condition of the target organ based on the search results.

[0034] For example, the depth feature of the image of the organ to be detected is encoded by a local sensitive hash function, and after encoding, a proximity search is performed in a Bloom filter to evaluate the pathological condition of the target organ according to the search result, wherein the local sensitive hash function h is represented by the following formula:

[0035]

[0036] In the above formula, w is a randomly generated normalized vector, b is the offset of random sampling, and ε is a hyperparameter used to control the width of the encoding interval.

[0037] For example, the pathological condition is combined with the target organ image as auxiliary diagnosis information, and combined with the preset interface layout principle for visual display, including:

[0038] Based on the visual attention relevance principle and the objective operation principle, the interface layout principle is constructed and transformed into a computable multi-objective optimization function of the interface layout;

[0039] The target optimization function is encoded through a preset algorithm to seek the optimal spatial solution, and then the optimal solution for the layout of the combined interface of auxiliary diagnostic information and target organ images is obtained, and a visual display is performed based on the optimal solution.

[0040] A second aspect of an embodiment of the present application provides an endoscope display enhancement system, comprising:

[0041] The enhancement module is used to obtain the original image of the endoscope and process the original image according to the preset algorithm to obtain the enhanced image, specifically:

[0042] Stretching the corresponding channels of the endoscope raw image in the first color space;

[0043] Converting the stretched first color space image to a second color space, and stretching the corresponding channels in the second color space;

[0044] Converting the stretched second color space image to the first color space to obtain an enhanced image;

[0045] An extraction module, used for extracting a target organ image based on the enhanced image;

[0046] A detection module, used to detect the lesion area in the target organ image according to a preset recognition algorithm, and evaluate the lesion condition of the target organ based on the lesion area;

[0047] The display module is used to combine the lesion condition as auxiliary diagnosis information with the target organ image, and perform a visual display in combination with the preset interface layout principles.

[0048] A third aspect of an embodiment of the present application provides a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in the first aspect when executing the computer program.

[0049] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0050] A fifth aspect of the present application provides a computer program product, which, when executed on a terminal, enables the terminal to execute the steps of the method described in the first aspect.

[0051] As can be seen from the above, the embodiment of the present application enhances the original endoscope image, thereby suppressing noise, weakening or removing certain unnecessary information, and then highlighting the useful information in the image, thereby improving the visual effect of the endoscope image.

[0052] At the same time, the embodiment of the present application detects the lesion area in the target organ image according to the preset recognition algorithm, evaluates the lesion condition of the target organ based on the lesion area, and combines the lesion condition with the target organ image as auxiliary diagnostic information, combined with the preset interface layout principles, for visual display. Therefore, intelligent diagnosis can be performed, and adverse situations such as fatigue caused by frequent manual operations can be avoided. It can also ensure the consistency of the diagnosis results, thereby improving the reliability of the diagnosis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 It is a flow chart of an endoscope display enhancement method provided in an embodiment of the present application;

[0055] Figure 2 is a structural schematic diagram of an endoscope display enhancement system provided in an embodiment of the present application;

[0056] Figure 3 It is a structural diagram of a terminal provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.

[0058] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of an endoscope display enhancement method provided in an embodiment of the present application. Figure 1 As shown, an endoscope display enhancement method comprises the following steps:

[0059] Step 1, obtaining an original endoscope image, and processing the original image according to a preset algorithm to obtain an enhanced image;

[0060] Step 2, extracting the target organ image based on the enhanced image;

[0061] Step 3, detecting the lesion area in the target organ image according to a preset recognition algorithm, and evaluating the lesion condition of the target organ based on the lesion area;

[0062] Step 4: Combine the lesion condition as auxiliary diagnosis information with the target organ image and perform a visual display in accordance with the preset interface layout principles.

[0063] Obviously, since the embodiment of the present application performs enhancement processing on the original endoscope image, it is possible to suppress noise, weaken or remove certain unnecessary information, and then highlight the useful information in the image, so that the visual effect of the endoscope image is improved;

[0064] At the same time, since the lesion area in the target organ image is detected according to the preset recognition algorithm, and the lesion condition of the target organ is evaluated based on the lesion area, and the lesion condition is combined with the target organ image as auxiliary diagnostic information, combined with the preset interface layout principles, it is visualized. Therefore, intelligent diagnosis can be performed, and adverse situations such as fatigue caused by frequent manual operations can be avoided. It can also ensure the consistency of the diagnosis results, thereby improving the reliability of the diagnosis results.

[0065] In the embodiment of the present application, step 1, obtaining an original endoscope image, and processing the original image according to a preset algorithm to obtain an enhanced image, may include the following contents:

[0066] Stretching the corresponding channel of the original endoscope image in the first color space, for example, stretching the grayscale contrast of the corresponding channel;

[0067] Converting the stretched first color space image to a second color space, and stretching the corresponding channel in the second color space, for example, stretching the grayscale contrast of the corresponding channel;

[0068] The stretched second color space image is converted to the first color space to obtain an enhanced image.

[0069] Due to interference from various factors, the contrast between organs and background in the original endoscopic image is low. In the embodiment of the present application, the original endoscopic image is stretched in the channels of the first color space and the second color space in sequence to enhance the contrast. Therefore, the noise interference of the image can be effectively reduced, thereby obtaining an image with better visual effect, that is, an enhanced image, to provide clear and reliable image data for disease prevention, analysis and diagnosis.

[0070] Specifically, the corresponding channel of the endoscope original image is stretched in the first color space, wherein the first color space may be an RGB color space, and the mapping function for contrast stretching the green (G) channel in the RGB color space may be expressed by the following formula:

[0071] g * (x,y)=[1+d g ×((g(x,y)) -1 -1) 2 ] -1

[0072] In the above formula, g(x, y) and g * (x, y) are the normalized grayscale values ​​of the G channel of the original image and the enhanced image at the pixel point (x, y); d g It is the grayscale mapping parameter of G channel.

[0073] In addition, the contrast stretching amounts of the red (R) channel and the blue (B) channel may be adjusted or determined based on the green (G) channel, which is not limited here.

[0074] Specifically, the stretched first color space image is converted to the second color space, and the corresponding channel is stretched in the second color space, wherein the second color space may be an HSV color space, and the mapping function for stretching the grayscale contrast of the saturation (S) channel in the second color space may be expressed by the following formula:

[0075] s * =[1+d s ×(s -1 -1) 2 ] -1

[0076] In the above formula, s and s * are the grayscale values ​​of the S channel before and after mapping; d s It is the S channel grayscale mapping parameter.

[0077] In the embodiment of the present application, step 2, extracting the target organ image based on the enhanced image, may include the following contents:

[0078] The enhanced image is subjected to superpixel gridding processing by a preset segmentation algorithm to obtain a plurality of superpixels, wherein the preset segmentation algorithm may be a SLIC superpixel segmentation algorithm;

[0079] Superpixels are classified based on convolutional neural networks to distinguish between superpixels inside organs and superpixels at the edge of organs.

[0080] Select several seed points of organ edge superpixels to construct the initial contour;

[0081] Based on the initial contour, the target organ image is segmented and extracted by solving the minimum value of the preset function.

[0082] Obviously, since the embodiment of the present application distinguishes between superpixels inside the organ and superpixels at the edge of the organ through a convolutional neural network, and constructs an initial contour by selecting seed points of several superpixels at the edge of the organ, and then segments the target organ image and extracts it by combining the initial contour with the solution of the minimum value of the preset function, it can reduce the sensitivity of the initial contour, reduce the segmentation time, and improve the accuracy of the segmentation results, thereby ensuring the integrity and accuracy of the extracted organs.

[0083] Specifically, superpixels are classified based on a convolutional neural network to distinguish between superpixels inside an organ and superpixels at the edge of an organ, wherein the objective function of the convolutional neural network can be:

[0084]

[0085] In the above formula, represents the probability distribution of the pixel point x output by the soft-max layer of the convolutional neural network, i represents the number of layers of the convolutional neural network, λi is the weight factor of the i-th layer, N is the total number of pixels, Ni is the total number of pixels in a category Lm, k represents the basic fact category label, and k∈[0,1] represents the basic fact category label, where 0 represents the superpixel inside the organ, 1 represents the superpixel at the edge of the organ, and x is the pixel point.

[0086] In this embodiment, the convolutional neural network can be trained by a stochastic gradient descent method.

[0087] In the embodiment of the present application, step 3, according to a preset recognition algorithm, detects the lesion area in the target organ image, and evaluates the lesion condition of the target organ according to the lesion area, which can be specifically:

[0088] Acquire a number of lesion images, and extract the depth features of the lesion images one by one;

[0089] The deep features of each lesion image are encoded into multiple groups of integers through a hash algorithm and stored in a Bloom filter;

[0090] When receiving the image of the target organ to be detected, extracting the depth feature of the image of the organ to be detected;

[0091] The deep features of the image of the organ to be detected are encoded by a local sensitive hash function, and after encoding, a proximity search is performed in a Bloom filter to evaluate the lesion condition of the target organ based on the search results. For example, when no approximation is retrieved, it means that there is no lesion in the image of the target organ. When the approximation is retrieved, the lesion condition is output based on the approximate lesion image.

[0092] Obviously, since the deep features of several lesion images are encoded as integers and stored in the Bloom filter, when the target organ image needs to be detected, it is only necessary to encode the deep features of the target organ image using a local sensitive hash function and perform a proximity search in the Bloom filter. Therefore, the retrieval results can be output quickly, thereby not only reducing latency but also having a very high throughput.

[0093] Specifically, the deep features of the image of the organ to be detected are encoded by a local sensitive hash function, and after encoding, a proximity search is performed in the Bloom filter to evaluate the pathological condition of the target organ according to the search result, wherein the local sensitive hash function h is a type of hash function that encodes a floating point vector into an integer or Boolean vector, which can be expressed by the following formula:

[0094]

[0095] In the above formula, w is a randomly generated normalized vector, b is the offset of random sampling, and ε is a hyperparameter used to control the width of the encoding interval;

[0096] Obviously, since the local sensitive hash function is used to encode the deep features of the image of the organ to be detected, it is very easy to encode it into the same integer, which is very convenient for retrieval, thereby improving computing efficiency and saving costs.

[0097] In the embodiment of the present application, step 4, combining the lesion condition as auxiliary diagnosis information with the target organ image, and combining with the preset interface layout principle for visual display, may include the following contents:

[0098] Based on the visual attention relevance principle and the objective operation principle, the interface layout principle is constructed and transformed into a computable multi-objective optimization function of the interface layout;

[0099] The target optimization function is encoded through a preset algorithm to seek the optimal spatial solution, and then the optimal solution for the layout of the combined interface of auxiliary diagnostic information and target organ images is obtained, and a visual display is performed based on the optimal solution.

[0100] It can be seen that the embodiment of the present application combines the visual attention correlation principle and the objective operation principle to construct an interface layout principle, and converts the interface layout principle into a computable interface layout multi-objective optimization function, and then obtains the optimal solution for the interface layout by solving the objective optimization function. Obviously, the visual display of auxiliary diagnostic information and target organ images based on the optimal solution can effectively enhance the visual effect and thus improve the user experience.

[0101] Based on the same inventive concept, the present application embodiment provides an endoscope display enhancement system, such as Figure 2 As shown, including:

[0102] Enhancement module 1 is used to obtain the original image of the endoscope, and process the original image according to a preset algorithm to obtain an enhanced image, specifically:

[0103] Stretching the corresponding channels of the endoscope raw image in the first color space;

[0104] Converting the stretched first color space image to a second color space, and stretching the corresponding channels in the second color space;

[0105] Converting the stretched second color space image to the first color space to obtain an enhanced image;

[0106] Extraction module 2, used for extracting the target organ image based on the enhanced image;

[0107] Detection module 3, used to detect the lesion area in the target organ image according to a preset recognition algorithm, and evaluate the lesion condition of the target organ based on the lesion area;

[0108] The display module 4 is used to combine the lesion condition as auxiliary diagnosis information with the target organ image and perform visual display in combination with the preset interface layout principle.

[0109] Obviously, since the embodiment of the present application performs enhancement processing on the original endoscope image, it is possible to suppress noise, weaken or remove certain unnecessary information, and then highlight the useful information in the image, so that the visual effect of the endoscope image is improved;

[0110] At the same time, since the lesion area in the target organ image is detected according to the preset recognition algorithm, and the lesion condition of the target organ is evaluated based on the lesion area, and the lesion condition is combined with the target organ image as auxiliary diagnostic information, combined with the preset interface layout principles, it is visualized. Therefore, intelligent diagnosis can be performed, and adverse situations such as fatigue caused by frequent manual operations can be avoided. It can also ensure the consistency of the diagnosis results, thereby improving the reliability of the diagnosis results.

[0111] Figure 3 is a structural diagram of a terminal provided in an embodiment of the present application. As shown in the figure, the terminal 3 of the embodiment includes: at least one processor 30 ( Figure 3 Only one is shown in the figure), a memory 31, and a computer program 32 stored in the memory 31 and executable on the at least one processor 30, wherein the processor 30 implements the steps of any of the above-mentioned method embodiments when executing the computer program 32.

[0112] The terminal 3 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will appreciate that Figure 3 It is only an example of terminal 3 and does not constitute a limitation on terminal 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0113] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0114] The memory 31 may be an internal storage unit of the terminal 3, such as a hard disk or memory of the terminal 3. The memory 31 may also be an external storage device of the terminal 3, such as a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), etc. equipped on the terminal 3. Further, the memory 31 may also include both an internal storage unit of the terminal 3 and an external storage device. The memory 31 is used to store the computer program and other programs and data required by the terminal. The memory 31 may also be used to temporarily store data that has been output or is to be output.

[0115] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for enhancing endoscope display, characterized in that: include: Acquire the original image of the endoscope, and process the original image according to a preset algorithm to obtain an enhanced image; Extracting the target organ image based on the enhanced image; According to the preset recognition algorithm, the lesion area in the target organ image is detected, and the lesion condition of the target organ is evaluated based on the lesion area; Combine the lesion condition as auxiliary diagnosis information with the target organ image and perform a visual display based on the preset interface layout principles; The method of obtaining an original endoscope image and processing the original image according to a preset algorithm to obtain an enhanced image includes: Stretching the corresponding channels of the endoscope raw image in the first color space; Converting the stretched first color space image to a second color space, and stretching the corresponding channels in the second color space; The stretched second color space image is converted to the first color space to obtain an enhanced image.

2. The endoscope display enhancement method according to claim 1, characterized in that: The corresponding channels of the endoscope original image are stretched in the first color space, wherein the first color space is the RGB color space, and the mapping function for contrast stretching the green channel in the RGB color space is expressed by the following formula: g * (x,y)=[1+d g ×((g(x,y)) -1 -1) 2 ] -1 In the above formula, g(x, y) and g * (x, y) are the normalized grayscale values ​​of the green channel of the original image and the enhanced image at the pixel point (x, y); d g It is the green channel grayscale mapping parameter.

3. The endoscope display enhancement method according to claim 1, characterized in that: The stretched first color space image is converted to the second color space, and the corresponding channel is stretched in the second color space, wherein the second color space is the HSV color space, and the mapping function for stretching the grayscale contrast of the saturation channel in the second color space is expressed by the following formula: s * =[1+d s ×(s -1 -1) 2 ] -1 In the above formula, s and s * are the grayscale values ​​of the saturation channel before and after mapping; d s It is the S channel grayscale mapping parameter.

4. The endoscope display enhancement method according to claim 1, characterized in that: The step of extracting the target organ image based on the enhanced image comprises: The enhanced image is processed into superpixels by a preset segmentation algorithm to obtain a number of superpixels; Superpixels are classified based on convolutional neural networks to distinguish between superpixels inside organs and superpixels at the edge of organs. Select several seed points of organ edge superpixels to construct the initial contour; Based on the initial contour, the target organ image is segmented and extracted by solving the minimum value of the preset function.

5. The endoscope display enhancement method according to claim 4, characterized in that: The superpixels are classified based on the convolutional neural network to distinguish the superpixels inside the organ and the superpixels at the edge of the organ, wherein the objective function of the convolutional neural network is: In the above formula, represents the probability distribution of the pixel x output by the soft-max layer of the convolutional neural network, i represents the number of layers of the convolutional neural network, and λ i is the weight factor of the i-th layer, N is the total number of pixels, N i is a category L n , k represents the basic fact category label, and k∈[0,1] represents the basic fact category label, where 0 represents the superpixel inside the organ, 1 represents the superpixel at the edge of the organ, and x is the pixel point.

6. The endoscope display enhancement method according to claim 1, characterized in that: The method of detecting the lesion area in the target organ image according to the preset recognition algorithm and evaluating the lesion condition of the target organ according to the lesion area is specifically as follows: Acquire a number of lesion images, and extract the depth features of the lesion images one by one; The deep features of each lesion image are encoded into multiple groups of integers through a hash algorithm and stored in a Bloom filter; When receiving the image of the target organ to be detected, extracting the depth feature of the image of the organ to be detected; The deep features of the image of the organ to be detected are encoded by a local sensitive hash function, and then a proximity search is performed in a Bloom filter to evaluate the pathological condition of the target organ based on the search results.

7. The endoscope display enhancement method according to claim 6, characterized in that: The depth feature of the image of the organ to be detected is encoded by a local sensitive hash function, and after encoding, a proximity search is performed in a Bloom filter to evaluate the pathological condition of the target organ according to the search result, wherein the local sensitive hash function h is represented by the following formula: In the above formula, w is a randomly generated normalized vector, b is the offset of random sampling, and ε is a hyperparameter used to control the width of the encoding interval.

8. The endoscope display enhancement method according to claim 1, characterized in that: The lesion condition is combined with the target organ image as auxiliary diagnosis information and visualized in combination with the preset interface layout principle, including: Construct interface layout principles and transform them into computable interface layout multi-objective optimization functions; The target optimization function is encoded through a preset algorithm to seek the optimal spatial solution, and then the optimal solution for the layout of the combined interface of auxiliary diagnostic information and target organ images is obtained, and a visual display is performed based on the optimal solution.

9. The endoscope display enhancement method according to claim 8, characterized in that: The method of constructing interface layout principles and converting the interface layout principles into a computable interface layout multi-objective optimization function is specifically as follows: based on the visual attention correlation principle and the objective operation principle, the interface layout principles are constructed and converted into a computable interface layout multi-objective optimization function.

10. An endoscope display enhancement system, the system implementing the method according to claim 1, characterized in that: include: The enhancement module is used to obtain the original image of the endoscope and process the original image according to the preset algorithm to obtain the enhanced image, specifically: Stretching the corresponding channels of the endoscope raw image in the first color space; Converting the stretched first color space image to a second color space, and stretching the corresponding channels in the second color space; Converting the stretched second color space image to the first color space to obtain an enhanced image; An extraction module, used for extracting a target organ image based on the enhanced image; A detection module, used to detect the lesion area in the target organ image according to a preset recognition algorithm, and evaluate the lesion condition of the target organ based on the lesion area; The display module is used to combine the lesion condition as auxiliary diagnosis information with the target organ image, and perform a visual display in combination with the preset interface layout principles.

Images