An image recognition system and method for wireless endoscopy

Abstract

The application discloses an image recognition system for wireless endoscopes, and relates to the technical field of wireless endoscopes, which comprises an image acquisition module, an image processing module, an AI algorithm end and a display module; the image acquisition module is used for acquiring image data in a real-time manner during an endoscopic surgery process, and transmits the real-time image data to a central processing unit through wired connection; the image processing module is used for processing the image; the image processing module is used for pre-processing and depth processing of the acquired image data; the pre-processing comprises image denoising, and the image denoising adopts a median filter algorithm; the depth processing is processed by using an AI algorithm, and comprises image graying, contrast enhancement, color enhancement and image segmentation; the image data is subjected to image graying, contrast enhancement, color enhancement and image segmentation processing in a respective manner, high-definition image data can be obtained after processing, the surgery picture display is very clear, doctors can make accurate judgments, and the quality of the surgery is improved.

Description

Technical Field

[0001] This invention relates to the field of wireless endoscope technology, and in particular to an image recognition system and method for wireless endoscopes. Background Technology

[0002] The Department of Urology is a hospital department that mainly diagnoses and treats certain diseases of the urinary system. It primarily treats various urological diseases. With the continuous development of the medical industry, 5G wireless endoscopy is gradually being applied in the clinical treatment of urological diseases and has achieved good therapeutic results.

[0003] In existing technologies, wireless endoscopes generally consist of a light source system, a power supply system, and an image recognition system. The image recognition system captures images during surgery through a camera, processes the images, and displays them on a screen, allowing doctors to perform precise operations based on the surgical images.

[0004] However, the image recognition system of the aforementioned wireless endoscope only performs simple image processing, which can easily lead to unclear image display and make it impossible to observe many details during the operation in a timely manner, thus affecting the quality of the operation. Therefore, there is room for improvement. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an image recognition system and method for wireless endoscopes. Its advantages lie in the fact that image data undergoes separate processing including grayscale conversion, contrast enhancement, color enhancement, and image segmentation. This processing yields high-definition image data, resulting in a very clear surgical view, facilitating accurate judgment by the surgeon and improving the quality of the surgery.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] An image recognition system and method for wireless laparoscopy includes an image acquisition module, an image processing module, an AI algorithm terminal, and a display module. The image acquisition module acquires image data in real time during laparoscopic surgery and transmits the real-time image data to a central processing unit via a wired connection. The image processing module processes the images, performing preprocessing and depth processing on the acquired image data. The preprocessing includes image denoising using a median filtering algorithm. The depth processing utilizes an AI algorithm, including image grayscale conversion, contrast enhancement, color enhancement, and image segmentation. The processed high-definition image data is then wirelessly transmitted to the display module, which displays the surgical footage.

[0008] The present invention is further configured such that the median filtering algorithm sorts the gray values ​​of all pixels in the pixel and its neighborhood window, takes the median as the new gray value of the pixel, replaces the original value with the median as the new gray value, so that the gray values ​​around it are close to the actual value, thereby eliminating noise points.

[0009] The present invention is further configured such that after the image is smoothed by the median filtering algorithm, some details will be blurred. The Laplacian operator is used to sharpen the image, laying the foundation for image segmentation.

[0010] The present invention is further configured such that the image grayscale processing quantizes the brightness value of the image from pure black to pure white into 0-255, a total of 256 levels. The grayscale value calculation formula is: g=(0.299R+0.578G+0.114B) / 3, where R, G and B represent the three primary colors corresponding to the brightness of each pixel in the image, and the grayscale value is obtained by weighted averaging of the three values.

[0011] The present invention is further configured such that the image segmentation employs an image segmentation function, which is: f(x,y) represents the initial image function, and T represents the set threshold.

[0012] The present invention is further configured such that the image is calculated using an image segmentation function, and the pixels are marked as objects or backgrounds accordingly to achieve image segmentation; however, using a fixed grayscale threshold may result in ineffective image segmentation due to factors such as uneven light brightness; therefore, it is necessary to subdivide the image into several sub-images, perform a weighted average of the grayscale values ​​of the neighboring pixels of a certain pixel in the sub-image to obtain the threshold of that point, and further obtain the average threshold of the sub-images as the image segmentation threshold.

[0013] The present invention is further configured such that the image contrast enhancement includes histogram equalization, contrast stretching, local contrast enhancement, adaptive histogram equalization, and nonlinear contrast enhancement.

[0014] A method of using an image recognition system for a wireless endoscope includes the following steps:

[0015] Step 1: First, the image acquisition module acquires image data in real time during the laparoscopic surgery process, and transmits the real-time image data to the central processing unit via a wired connection to process the images using the image processing module;

[0016] Step 2: The image processing module performs preprocessing and depth processing on the image data. First, it sorts the gray values ​​of all pixels in the pixel and its neighborhood window, takes the median value as the new gray value of the pixel, and replaces the original value with the median value as the new gray value, so that the gray values ​​of the surrounding pixels are close to the actual values, thereby eliminating noise points.

[0017] Step 3: Next, AI algorithms are used to process the image data by converting it to grayscale, enhancing contrast, enhancing color, and segmenting the image. After processing, high-definition image data can be obtained. Finally, the surgical screen is displayed using a display module, allowing the doctor to make accurate judgments based on the screen during the surgery.

[0018] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0019] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0020] The beneficial effects of this invention are as follows: This invention discloses an image recognition system for wireless endoscopes. After the wireless endoscope acquires images, the image data undergoes preprocessing and depth processing. First, the grayscale values ​​of all pixels within a pixel and its neighborhood window are sorted, and the median value is taken as the new grayscale value of that pixel. The median value replaces the original value as the new grayscale value, making the surrounding grayscale values ​​close to the actual values, thereby eliminating noise points. Using AI algorithms, the image data is processed for image grayscale conversion, contrast enhancement, color enhancement, and image segmentation. After processing, high-definition image data can be obtained, and the surgical screen is displayed very clearly, facilitating accurate judgment by doctors and improving the quality of surgery. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall structure of an image recognition system for a wireless endoscope proposed in this invention;

[0022] Figure 2 This is a schematic diagram illustrating the workflow of an image recognition system for a wireless endoscope proposed in this invention. Detailed Implementation

[0023] The technical solution of this patent will be further described in detail below with reference to specific embodiments.

[0024] The embodiments of this patent are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this patent, and should not be construed as limiting this patent.

[0025] Reference Figure 1An image recognition system for wireless endoscopy includes an image acquisition module, an image processing module, an AI algorithm module, and a display module. The image acquisition module acquires image data in real time during laparoscopic surgery and transmits the real-time image data to a central processing unit via a wired connection. The image processing module processes the images, performing preprocessing and depth processing on the acquired image data. Preprocessing includes image denoising using a median filtering algorithm. Depth processing utilizes AI algorithms, including image grayscale conversion, contrast enhancement, color enhancement, and image segmentation. The processed high-definition image data is then wirelessly transmitted to the display module, which displays the surgical images.

[0026] In this embodiment, the median filtering algorithm sorts the gray values ​​of all pixels in the pixel and its neighborhood window, takes the median as the new gray value of the pixel, and replaces the original value with the median as the new gray value, so that the gray values ​​of the surrounding pixels are close to the actual values, thereby eliminating noise points. After the image is smoothed by the median filtering algorithm, some details will be blurred. The Laplacian operator is used to sharpen the image, laying the foundation for image segmentation.

[0027] Furthermore, image segmentation is the most critical step in image processing, and grayscale thresholding segmentation algorithms are widely used because of their fast computation speed and low computational cost.

[0028] Image grayscale processing quantizes the brightness values ​​of an image from pure black to pure white into 0-255, a total of 256 levels. The grayscale value calculation formula is: g=(0.299R+0.578G+0.114B) / 3, where R, G and B represent the three primary colors corresponding to the brightness of each pixel in the image. The grayscale value is obtained by weighted averaging of the three values.

[0029] Image segmentation uses an image segmentation function, which is:

[0030] f(x,y) represents the initial image function, and T represents the set threshold. The image is segmented using the image segmentation function, and the pixels are marked as objects or backgrounds accordingly. However, using a fixed grayscale threshold may not effectively segment the image due to factors such as uneven lighting. Therefore, the image needs to be subdivided into several sub-images. The grayscale values ​​of the neighboring pixels of a certain pixel in the sub-image are weighted and averaged to obtain the threshold of that point. The average threshold of the sub-images is then obtained as the segmentation threshold of the image.

[0031] It is worth mentioning that image contrast enhancement includes histogram equalization, contrast stretching, local contrast enhancement, adaptive histogram equalization, and nonlinear contrast enhancement.

[0032] Histogram equalization redistributes the gray values ​​of an image, making the gray values ​​of the output image roughly uniform, thereby enhancing contrast; it is applicable to both grayscale and color images and does not require specifying parameters;

[0033] Adaptive histogram equalization divides the image into several small regions and performs histogram equalization on each region separately to preserve local contrast.

[0034] Contrast stretching, also known as grayscale stretching, is an image enhancement technique. It improves image quality by changing the brightness values ​​of image pixels, thereby increasing the contrast of the entire image or specific areas.

[0035] Main idea: Improve the dynamic range of gray levels during image processing to make image details clearer.

[0036] Implementation method: Piecewise linear transformation functions are typically used to map a specific grayscale range of an image to a new dynamic range.

[0037] Effect: It allows selective stretching of a specific grayscale range to improve the output image. If the image's grayscale is concentrated in darker or brighter areas, image quality can be improved by stretching or compressing the corresponding grayscale range.

[0038] Nonlinear contrast enhancement is an image enhancement technique that addresses the problem of concentrated gray-level distribution and low contrast in images. It mainly includes the following methods: exponential transform, logarithmic transform, and gamma transform. These methods transform the image gray levels using nonlinear functions, thereby expanding or compressing the gray levels and improving image contrast. For example, the gamma transform changes the image brightness by adjusting the gamma value; when gamma < 1, low-brightness pixels are enhanced, and when gamma > 1, high-brightness pixels are reduced.

[0039] Reference Figure 2 A method of using an image recognition system for a wireless endoscope includes the following steps:

[0040] Step 1: First, the image acquisition module acquires image data in real time during the laparoscopic surgery process, and transmits the real-time image data to the central processing unit via a wired connection to process the images using the image processing module;

[0041] Step 2: The image processing module performs preprocessing and depth processing on the image data. First, it sorts the gray values ​​of all pixels in the pixel and its neighborhood window, takes the median value as the new gray value of the pixel, and replaces the original value with the median value as the new gray value, so that the gray values ​​of the surrounding pixels are close to the actual values, thereby eliminating noise points.

[0042] Step 3: Next, AI algorithms are used to process the image data by converting it to grayscale, enhancing contrast, enhancing color, and segmenting the image. After processing, high-definition image data is obtained. Finally, the surgical screen is displayed using a display module. During the surgery, the doctor makes accurate judgments based on the screen. The high-definition image data obtained after processing makes the surgical screen very clear, improving the quality of the surgery.

[0043] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0044] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above method.

[0045] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An image recognition system for a wireless endoscope, characterized in that, It includes an image acquisition module, an image processing module, an AI algorithm module, and a display module; The image acquisition module is used to acquire image data in real time during laparoscopic surgery. The real-time image data is transmitted to the central processing unit via a wired connection, and the image processing module processes the image. The image processing module performs preprocessing and depth processing on the acquired image data. The preprocessing includes image denoising, which uses a median filtering algorithm. The deep processing utilizes AI algorithms, including image grayscale conversion, contrast enhancement, color enhancement, and image segmentation. The processed high-definition image data is then wirelessly transmitted to the display module, which displays the surgical footage.

2. The image recognition system for a wireless endoscope according to claim 1, characterized in that, The median filtering algorithm sorts the gray values ​​of all pixels in the pixel and its neighborhood window, takes the median as the new gray value of the pixel, replaces the original value with the median as the new gray value, and makes the gray values ​​of the surrounding pixels closer to the actual value, thereby eliminating noise points.

3. The image recognition system for a wireless endoscope according to claim 2, characterized in that, After the image is smoothed by the median filtering algorithm, some details will be blurred. The Laplacian operator is used to sharpen the image, laying the foundation for image segmentation.

4. The image recognition system for a wireless endoscope according to claim 1, characterized in that, The image grayscale processing quantizes the brightness values ​​of the image from pure black to pure white into 0-255, a total of 256 levels. The grayscale value calculation formula is: g=(0.299R+0.578G+0.114B) / 3, where R, G and B represent the three primary colors corresponding to the brightness of each pixel in the image, and the grayscale value is obtained by weighted averaging of the three values.

5. The image recognition system for a wireless endoscope according to claim 1, characterized in that, The image segmentation uses an image segmentation function, which is: f(x,y) represents the initial image function, and T represents the set threshold.

6. The image recognition system for a wireless endoscope according to claim 5, characterized in that, The image is segmented using an image segmentation function, and the pixels are marked as objects or backgrounds accordingly to achieve image segmentation. However, using a fixed grayscale threshold may not effectively segment the image due to factors such as uneven light intensity. Therefore, the image needs to be subdivided into several sub-images, and the grayscale values ​​of the neighboring pixels of a certain pixel in the sub-image are weighted and averaged to obtain the threshold of that point. The average threshold of the sub-image is then obtained as the image segmentation threshold.

7. The image recognition system for a wireless endoscope according to claim 1, characterized in that, The image contrast enhancement includes histogram equalization, contrast stretching, local contrast enhancement, adaptive histogram equalization, and nonlinear contrast enhancement.

8. A method of using an image recognition system for a wireless endoscope as described in any one of claims 1-7, characterized in that, Includes the following steps: Step 1: First, the image acquisition module acquires image data in real time during the laparoscopic surgery process, and transmits the real-time image data to the central processing unit via a wired connection to process the images using the image processing module; Step 2: The image processing module performs preprocessing and depth processing on the image data. First, it sorts the gray values ​​of all pixels in the pixel and its neighborhood window, takes the median value as the new gray value of the pixel, and replaces the original value with the median value as the new gray value, so that the gray values ​​of the surrounding pixels are close to the actual values, thereby eliminating noise points. Step 3: Next, AI algorithms are used to process the image data by converting it to grayscale, enhancing contrast, enhancing color, and segmenting the image. After processing, high-definition image data can be obtained. Finally, the surgical screen is displayed using a display module, allowing the doctor to make accurate judgments based on the screen during the surgery.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method of claim 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method described in claim 8.

Images