Specimen image generation method based on endoscope shooting and endoscope

Abstract

The invention provides a specimen image generation method based on endoscope shooting and an endoscope. The method comprises the following steps: acquiring a first category of images; wherein the first class image is used for indicating an image acquisition end of an endoscope to shoot an image acquired by a specimen in vitro; based on a preset distortion correction algorithm, distortion correction is carried out on the first type of images to obtain distortion correction images, and the distortion correction algorithm is an algorithm used for carrying out distortion correction on the first type of images; and storing the distortion correction image as a specimen image. By means of the method, the hardware cost can be effectively reduced, the problems that in the treatment process of organ or tissue specimens cut by an endoscope, the flow is complex, and shooting is delayed are solved, meanwhile, distortion correction is conducted on the in-vitro shot images, the images can be displayed according to the normal proportion, and the observation and recording effects of a user are improved.

Description

Technical Field

[0001] The present application relates to the technical field of endoscopic image processing, and in particular to a method for generating a specimen image based on endoscopic photography and an endoscope. Background Art

[0002] An endoscope is an instrument used in medical examination and surgery, designed to enter the human body through natural openings or small incisions to directly observe and remove diseased organs or tissues inside the body.

[0003] When using an endoscope to remove diseased organs or tissues inside the human body, such as during endoscopic submucosal dissection (ESD), medical staff need to use the removed diseased organs or tissues to further evaluate whether the disease has been cured in order to determine whether additional treatment is needed. To make these evaluations, it is necessary to use specimen images of the removed diseased organs or tissues for pathological diagnosis.

[0004] Currently, specimen images are primarily captured using cameras or other imaging equipment to capture the diseased organs or tissues being removed. Using additional imaging equipment increases hardware costs. Furthermore, in medical settings such as endoscopic submucosal dissection procedures, the endoscope must be replaced with another imaging device to capture the specimen, leading to complex workflows, delays in capture, and the need for additional image output. Summary of the Invention

[0005] The present application provides a method for generating a specimen image based on endoscope photography and an endoscope to solve at least one of the above problems.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] The present application provides an image processing method for endoscopic images, comprising:

[0008] Acquire a first category of images; wherein the first category of images is used to indicate images acquired by an image acquisition end of an endoscope photographing a specimen in vitro;

[0009] Based on a preset distortion correction algorithm, performing distortion correction on the first category of images to obtain a distortion-corrected image, wherein the distortion correction algorithm is an algorithm for performing distortion correction on the first category of images;

[0010] The distortion-corrected image is saved as a specimen image.

[0011] In one embodiment, acquiring the first category image includes:

[0012] Acquiring an initial image captured by an image acquisition terminal of the endoscope;

[0013] According to at least one of the image information, identification information and response information of the initial image, it is determined that the initial image is an image of the first category.

[0014] In one embodiment, determining, based on the image information of the initial image, that the initial image is an image of the first category includes:

[0015] Determining target pixels of the initial image, wherein the target pixels are at least a portion of pixels in the initial image after brightness filtering;

[0016] Determining the hemoglobin index corresponding to the target pixel according to the channel values ​​of the red and blue channels corresponding to the target pixel;

[0017] determining image information of the initial image according to the hemoglobin index corresponding to the target pixel;

[0018] And / or, determining, based on the manual recognition information of the initial image, that the initial image is an image of the first category includes:

[0019] In response to operation information on the initial image, acquiring human recognition information about the initial image;

[0020] Response information of the initial image is obtained according to the manual recognition information.

[0021] In one embodiment, determining the image information of the initial image according to the hemoglobin index corresponding to the target pixel includes:

[0022] determining, in the initial image, the number of target pixels whose hemoglobin index reaches the preset index threshold according to the hemoglobin index corresponding to the target pixel and a preset index threshold;

[0023] determining a ratio based on the number and the total number of target pixels in the initial image;

[0024] According to the ratio and a preset ratio threshold, the initial image is determined to be an image of the first category.

[0025] In one embodiment, determining the target pixel of the initial image includes:

[0026] For each pixel in the initial image, determining a brightness value corresponding to each pixel according to the sum of a channel value of the red channel weighted by the first weight, a channel value of the green channel weighted by the second weight, and a channel value of the blue channel weighted by the third weight;

[0027] The target pixel of the initial image is determined according to the brightness value of each pixel and a preset brightness value.

[0028] In one embodiment, performing distortion correction on the first category image based on a preset distortion correction algorithm to obtain a distortion-corrected image includes:

[0029] Determining, according to the distortion correction algorithm, pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image;

[0030] According to the pixel coordinates corresponding to each pixel of the distortion-corrected image and the pixel values ​​corresponding to the pixel coordinates, each pixel of the distortion-corrected image is assigned a value one by one to obtain the distortion-corrected image.

[0031] In one embodiment, the distortion correction algorithm includes camera intrinsic parameters and distortion parameters; and determining, based on the distortion correction algorithm, pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image includes:

[0032] Determining the image plane coordinates of each pixel of the distortion-corrected image according to the pixel coordinates of each pixel of the distortion-corrected image and the inverse matrix of the camera intrinsic parameter;

[0033] According to the image plane coordinates of each pixel of the distortion-corrected image and the distortion parameter, pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image are determined.

[0034] In one embodiment, performing distortion correction on the first category image based on a preset distortion correction algorithm to obtain a distortion-corrected image includes:

[0035] Determining, according to the distortion correction algorithm, pixel coordinates corresponding to each pixel of the first category image in the distortion-corrected image;

[0036] According to the pixel coordinates corresponding to each pixel of the first category image and the pixel values ​​corresponding to the pixel coordinates, each pixel of the distortion-corrected image is assigned a value one by one to obtain the distortion-corrected image.

[0037] In one embodiment, after saving the distortion-corrected image as a specimen image, the method further includes:

[0038] sending the specimen image to a target terminal so that the target terminal stores or displays the specimen image; or

[0039] The specimen image is displayed.

[0040] In a second aspect, the present application provides an image processing device for endoscopic images, comprising:

[0041] A first acquisition module is configured to acquire a first category of images; wherein the first category of images is used to indicate images acquired by an image acquisition end of an endoscope photographing a specimen in vitro;

[0042] a correction module, configured to perform distortion correction on the first category of images based on a preset distortion correction algorithm to obtain a distortion-corrected image, wherein the distortion correction algorithm is an algorithm for performing distortion correction on the first category of images;

[0043] A saving module is used to save the distortion-corrected image as a specimen image.

[0044] In a third aspect, the present application provides an endoscope, comprising: a memory and a processor;

[0045] The memory stores computer-executable instructions;

[0046] The processor executes the computer-executable instructions stored in the memory, so that the endoscope executes the method for generating a specimen image based on endoscope photography.

[0047] In a fourth aspect, the present application provides a computer-readable storage medium comprising computer instructions. When the computer instructions are executed on a computer, the computer executes the specimen image generation method based on endoscopic photography provided in any one of the first aspects above.

[0048] In a fifth aspect, the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, it can implement the specimen image generation method based on endoscopic photography provided in any one of the above-mentioned first aspects.

[0049] The present embodiment provides a method for generating a specimen image based on endoscope photography and an endoscope. The method obtains a first category image, which is used to instruct the image acquisition end of the endoscope to capture an image of the specimen captured in vitro. The method performs distortion correction on the first category image based on a preset distortion correction algorithm to obtain a distortion-corrected image, and then saves the distortion-corrected image as the specimen image. In this process, the image captured by the specimen captured in vitro by the endoscope, i.e., the first category image, is corrected for distortion by the distortion correction algorithm and then saved. There is no need to use other camera equipment to capture the specimen image, which reduces hardware costs and solves problems such as complex processes and shooting delays. At the same time, considering that the first category image captured by the endoscope is prone to distortion, the first category image is corrected for distortion so that the specimen image can be displayed in normal proportion, which is convenient for subsequent users to better observe and record the specimen image. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1a Examples of organ or tissue specimens after resection;

[0052] Figure 1b A schematic diagram of a possible scenario provided in an embodiment of the present application;

[0053] Figure 2 A schematic flow chart of an image processing method for endoscopic images provided in an embodiment of the present application;

[0054] Figure 3a This is an example image without distortion;

[0055] Figure 3b This is an example of barrel distortion;

[0056] Figure 3c This is an example of pincushion distortion;

[0057] Figure 4 This is an example diagram of tangential distortion;

[0058] Figure 5 Example image of a checkerboard chart for endoscopy;

[0059] Figure 6a This is one of the display interface diagrams for the distortion-corrected image;

[0060] Figure 6b This is the second display interface diagram of the distortion-corrected image;

[0061] Figure 7a for Figure 2 Flow diagram of step S201;

[0062] Figure 7b One of the display interface diagrams for manual identification information;

[0063] Figure 7c This is the second display interface diagram for manual identification information;

[0064] Figure 8 A schematic diagram of the structure of an image processing device for endoscopic images provided in an embodiment of the present application;

[0065] Figure 9 A schematic diagram of the structure of the endoscope provided in an embodiment of the present application.

[0066] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0067] During endoscopic submucosal dissection (ESD), in order to determine whether additional treatment is needed or whether a radical cure is needed, a pathological diagnosis must be performed. Generally speaking, the diseased organ or tissue (i.e., specimen) after endoscopic resection should be quickly fixed on a table, such as a foam plastic, rubber board, or cork board. When fixing, place the mucosal surface upward, and the extent of the specimen should be consistent with the tumor observed under the endoscope. Fix it to the table with a pin, and then mark the oral and anal sides of the diseased tissue, such as Figure 1a As shown, the specimen after endoscopic resection can be processed as follows Figure 1a The method shown is fixed on the table, wherein Figure 1a The a in it represents a pin.

[0068] In the prior art, medical staff typically use cameras, cell phones, or other imaging devices to capture specimens that have been quickly fixed to a platen to produce an image. However, using separate imaging devices increases hardware costs. Furthermore, in medical settings such as endoscopic submucosal dissection procedures, the endoscope must be replaced with another imaging device to capture the specimen. This leads to an overly complex process, delayed capture, and the need for additional image output, which in turn affects the efficiency of specimen image processing.

[0069] The specimen image generation method and endoscope provided in this embodiment utilizes the endoscope's lens tip to align with a fixed specimen and capture a photograph for archiving. Since no additional camera equipment is required for capture, hardware costs are reduced, while also resolving issues such as complex processes and delayed capture. Furthermore, considering that specimen images captured by an endoscope are prone to distortion, distortion correction is performed on the specimen images to facilitate subsequent user observation and recommendations for further examination and treatment.

[0070] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below in conjunction with the drawings in the embodiments of the present application. In the drawings, the same or similar reference numerals throughout represent the same or similar parts or parts with the same or similar functions. The described embodiments are part of the embodiments of the present application, not all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present application, and should not be understood as limitations on the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0071] Figure 1b A possible scenario diagram provided for an embodiment of the present application is as follows: Figure 1b As shown, the terminal device 110 and the endoscope 120 are electrically connected. Medical personnel can use the endoscope 120 to collect images of tissues in the patient's body and diseased organs or tissue specimens after resection. The images of diseased organs or tissues after resection are called specimen images. The endoscope 120 collects the images of the specimens, corrects the distortion of the images of the specimens, and saves them. The images can be transmitted to the display screen of the terminal device 110 for display. Medical personnel can give suggestions for further examination and treatment by observing the specimen images. Alternatively, the terminal device 110 can store a specific specimen image analysis algorithm (such as a neural network model) to output the analysis results corresponding to the specimen images, etc. Optionally, the terminal device 110 may include, but is not limited to, a computer, a smart phone, a tablet computer, an e-book reader, a Moving Picture experts group audiolayer III (MP3) player, a Moving Picture experts group audio layer IV (MP4) player, a portable computer, a wearable device, a desktop computer, a set-top box, a smart TV, and the like.

[0072] Optionally, the electrical connection between the terminal device 110 and the endoscope 120 can be a wired connection, connecting the devices through physical cables such as data cables and network cables, such as Universal Serial Bus (USB), High-Definition Multimedia Interface (HDMI), Ethernet and other interfaces to connect devices for data transmission. It can also be a wireless connection, communicating through radio wave signals, such as Wireless Fidelity (WiFi), Bluetooth, infrared and other wireless technologies to achieve communication between devices. Or it can be a remote connection, connecting devices through network technologies such as the Internet to achieve remote control and data transmission, such as cloud services, remote desktops, etc. It can also be a near field communication (NFC) connection, connecting devices through near field communication technology, which is usually used for applications such as file transfer; low-power Bluetooth connection, used for short-range communication between devices, such as the connection of devices such as smart bracelets and smart watches; wireless communication technology connection, used for low-speed, short-range communication between devices, suitable for low-power Internet of Things devices. This application does not specifically limit the specific method of communication connection between physical devices.

[0073] The above is a brief description of the scenario diagram of this application. Figure 1b The endoscope 120 in the embodiment of the present application is taken as an example to explain in detail the specimen image generation method based on endoscope photography and the endoscope provided in the embodiment of the present application. In some embodiments, in addition to being applicable to endoscopes, the method can also be applied to other devices, such as the terminal device 110 or a server, etc.

[0074] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0075] Figure 2 A flowchart of an image processing method for endoscopic images is provided in an embodiment of the present application, as shown in FIG. Figure 2 As shown, the method may include steps S201-S203:

[0076] Step S201 : Acquire a first category image; wherein the first category image is used to indicate an image acquired by an image acquisition end of an endoscope photographing a specimen in vitro.

[0077] For example, the first category of images is an image corresponding to a specimen taken by an endoscope in vitro, so as to distinguish between in vivo images and in vitro images taken by the endoscope. The first category of images can be acquired in real time or stored in a storage space after acquisition.

[0078] Step S202: Based on a preset distortion correction algorithm, distortion correction is performed on the first category image to obtain a distortion-corrected image, wherein the distortion correction algorithm is an algorithm for performing distortion correction on the first category image.

[0079] Taking into account the different scenarios of in vivo and in vitro photography performed by the endoscope, the real-time tissue images taken in vivo may pay more attention to the clarity and detail of the real-time images to support immediate diagnosis, while the specimen images taken in vitro are more concerned with the geometric accuracy and consistency of the specimens, so that users can observe and analyze the specimens later. In addition, the distortions produced by in vivo and in vitro photography are also different. Using the distortion correction scheme for in vivo photography to process images taken in vitro may not be able to fully meet the high requirements of in vitro images for geometric accuracy and consistency, making it difficult to achieve the correction effect. Therefore, this embodiment uses a distortion correction algorithm dedicated to the first category of images to perform distortion correction on the first category of images, that is, the images taken in vitro, so that the geometric shape and size of the image are more consistent with the actual specimen, so that users can observe the specimens more easily.

[0080] Optionally, those skilled in the art can determine the preset distortion algorithm based on practical applications or prior data adaptability. For example, the distortion correction algorithm can be based on existing optical distortion models, which describe various types of distortion that may be introduced by the lens during the imaging process, such as radial distortion and tangential distortion, where radial distortion may cause barrel or pincushion distortion of the image, while tangential distortion may cause slight tilt or asymmetry of the image. A large number of in vitro images taken through endoscopy are collected for testing to calculate the various parameters in the distortion model, thereby determining the final distortion algorithm, which is then applied to the distortion correction process of the first category of images.

[0081] Exemplarily, a preset distortion correction algorithm may include distortion correction parameters and camera intrinsic parameters. The camera intrinsic parameters and distortion parameters may be calibrated by obtaining a preset number of sets of corner point coordinates and inputting these sets of corner point coordinates into a preset calibration algorithm model; wherein the preset number of sets of corner point coordinates are for the corner point coordinates of the target object image, and the calibration algorithm model includes the camera intrinsic parameters and distortion parameters to be solved; and then iteratively solving the calibration algorithm model using a nonlinear least squares method to obtain the distortion parameters and camera intrinsic parameters.

[0082] The geometric distortion introduced by the endoscope lens can be divided into two categories: 1) Radial distortion, because the light in different areas has different degrees of bending after passing through the lens, such as a small degree of bending in the center area and a large degree of bending in the boundary area, causing the real straight line to appear curved after imaging. Figure 3a-3b As shown in the figure, there are cases of no distortion, barrel distortion (Barrel Distortion) and pincushion distortion (Pincushion Distortion) in turn. 2) Tangential distortion (Tangential Distortion): Since the sensor and the lens are not completely parallel, the final image will be stretched in a certain direction. Figure 4 As shown, the dotted grid indicates no distortion, and the solid line indicates the presence of tangential distortion.

[0083] Research has shown that when endoscopes capture images outside the body, radial distortion is the primary cause. Due to the lens assembly process, some lenses also exhibit slight tangential distortion. To correct for this distortion, it is first necessary to obtain the parameters for radial and tangential distortion. These parameters can be calculated using a nonlinear least squares method by taking images of a standard chart (such as a checkerboard) from multiple angles.

[0084] Optionally, the method of obtaining a preset number of groups of corner point coordinates can be achieved by: capturing an image of the target object, where the target object image is at least a plurality of images taken in vitro by an endoscope at multiple angles and / or multiple distances of the target object; for each target object image, obtaining the corner point coordinates of the target object to obtain a predetermined preset number of groups of corner point coordinates; wherein the corner point coordinates include the physical coordinates and image coordinates of the corner point.

[0085] For example, the chessboard pattern card can be photographed with an endoscope from different angles and distances to obtain at least 20 photos. For example, a photo from one angle is as follows: Figure 5 As shown, the coordinate positions of each corner point (intersection of black and white squares) on the chart and in the image are taken, as shown in Table 1 below (coordinates of the first 20 corner points). The coordinates on the chart are called physical coordinates, with the origin being the first corner point in the upper left corner, the long side being the x-axis, the short side being the y-axis, and the direction perpendicular to the chart plane being the z-axis. The side length of each square is 5 mm. The coordinates in the image are called image coordinates, with the origin being the upper left corner of the image, the horizontal direction being the x-axis, the vertical direction being the y-axis, and the direction perpendicular to the image plane being the z-axis. The coordinate units are pixels.

[0086] Table 1

[0087]

[0088]

[0089] Combined with mathematical modeling of the camera imaging principle, for example, the distortion model can adopt the Brown-Conrady model, and the process of converting from physical coordinates to image coordinates can be obtained using the following formula, as shown in Table 2:

[0090] Table 2

[0091]

[0092] Where, (X w ,Y w ,Z w ) represents the physical coordinate, (X c ,Y c ,Z c ) represents the camera coordinates, (x, y, z) represents the pixel coordinates, (x', y') represents the phase plane coordinates, and (x", y") represents the phase plane coordinates after distortion; R represents the rotation matrix, and t represents the translation vector; r represents the radial distance, and the square of the radial distance is the square of the distance from the optical axis to the image plane coordinates, which is used to calculate the radial distortion; (k1, k2, k3) represents the radial distortion parameters, which describes the degree of radial distortion of the lens; (p1, p2): represents the tangential distortion parameters, which describes the degree of tangential distortion of the lens; γ is the radial distortion factor; (f x ,f y ), represents the focal length parameter in the camera intrinsic parameter, which represents the focal length in the horizontal and vertical directions respectively; (c x ,c y ) represents the optical center coordinates in the camera intrinsic parameters, indicating the position of the optical center relative to the upper left corner of the image; Radial and Tangential represent radial and tangential directions, respectively.

[0093] Among them, formula (1) converts the physical coordinates of the corner point into camera coordinates (with the lens as the coordinate origin, the xy plane is parallel to the plane where the lens is located, and the z-axis direction is the optical axis direction of the lens), which contains two parameters: the rotation matrix R and the translation vector t (the posture and position of the camera relative to the chessboard). These two parameters may be different for each image; formula (2) converts the camera coordinates of the corner point into image plane coordinates; formula (3) performs distortion transformation, converting the undistorted coordinate point (x′, y′) into the distorted coordinate point (x″, y″) through the distortion model, where k1, k2, k3 are radial distortion parameters, and p1 and p2 are tangential distortion parameters, all of which are unknown parameters. All images have the same specific distortion parameters.

[0094] Formula (4) converts image plane coordinates into image coordinates, where (c x ,c y ) represents the position of the optical center relative to the upper left corner of the image, f x and fy Represents the focal length in the horizontal and vertical directions. These parameters are related to the lens and are called camera intrinsic parameters. They are unknown parameters and are the same in all images.

[0095] Given the physical coordinates (X w ,Y w ,Z w ) and image coordinates (x, y, z), and the optimal parameters, including distortion parameters and camera intrinsics, can be iteratively solved using the nonlinear least squares method (Levenberg-Marquardt algorithm). As you can understand, nonlinear least squares is a parameter estimation method that estimates the parameters of a nonlinear static model by minimizing the sum of squared errors. This is an optimization technique for data fitting that aims to find a set of parameters that minimizes the sum of squared errors between the model predictions and the observed data.

[0096] Step S203: Save the distortion-corrected image as a specimen image.

[0097] For example, after obtaining the distortion-corrected image, it can be directly saved as a specimen image, or further interacted with the user to determine whether it meets the user's needs and then saved as a specimen image.

[0098] For example, the distortion-corrected image can be displayed on a display screen (if the execution subject is a terminal device, it can be displayed directly; if the execution subject is an endoscope, the distortion-corrected image can be sent to the terminal device for display and processing through interaction with the terminal device). After the user confirms the result through user interaction, the distortion-corrected image can be saved as a specimen image. In one implementation, the distortion-corrected image can be displayed on a display screen, such as Figure 6a As shown; the first category image before correction and the distortion-corrected image after correction can also be displayed on the display screen at the same time for on-screen comparison, which is convenient for users to observe, as shown Figure 6b shown.

[0099] In some embodiments, while displaying the distortion-corrected image (e.g., the terminal device obtains and displays the distortion-corrected image, or the endoscope sends the image to the terminal device for display), the terminal device may simultaneously pop up a confirmation box and / or a rejection box. The user triggers the confirmation box to save the distortion-corrected image, or automatically save it after a few seconds or automatically refuse to save it. If the user refuses to save, it means that the current distortion-corrected image does not meet the user's expectations. The first category image can be re-captured or the first category image can be re-distorted, and the above process can be executed until the user's needs are met, thereby further optimizing the user experience.

[0100] Through the above method, this embodiment reuses the endoscope's tip to align with the marked specimen and take a photo to archive the specimen image, reducing hardware costs and solving problems such as complex processes and shooting delays. At the same time, considering that specimen images taken by the endoscope are prone to distortion, distortion correction is performed on the specimen image to facilitate subsequent user observation and provide recommendations for further examination and treatment.

[0101] Taking into account the situation of non-real-time acquisition, since the endoscope can acquire in-vivo images and in-vitro images, or even more types of endoscopic images, this embodiment further refines the acquisition process of the first category of images. Figure 7a As shown, the above step S201 of acquiring the first category image can be divided into the following steps S2011 and S2012:

[0102] Step S2011: Acquire an initial image captured by the image acquisition end of the endoscope.

[0103] Optionally, the initial image may be an image captured by the endoscope inside the body or outside the body. The image captured by the endoscope outside the body may be an image captured for the specimen or may not be an image captured for the specimen.

[0104] Step S2012: Determine that the initial image is an image of the first category based on at least one of the image information, identification information, and response information of the initial image.

[0105] In one example, images captured by an endoscope are stored in a single storage space without distinguishing between them. In this case, the initial image can be combined with image information of the initial image to determine that the initial image is a first-category image. In another example, after capturing endoscopic images of different categories, identification information can be automatically added to the image. For example, an image captured of a specimen can be directly identified as a first-category image. By recognizing this identification information, the first-category image can be quickly determined.

[0106] Optionally, the identification information can be embedded in the pixels of the image using a digital watermark. Digital watermarks are usually invisible, and the identification information of the image can be obtained by reading the digital watermark carried by the image. Alternatively, a visible watermark can be used, such as adding visible identification information in a corner of the image, such as the date, time, or device ID. This method is simple and direct. In addition, metadata can be used, such as adding identification information to the EXIF ​​(Exchangeable Image File Format) data of the image file. This method does not change the image itself, but stores the information in the additional data part of the file. By reading the stored file, the identification information can be quickly obtained. In some embodiments, other methods can also be used to add identification information to the initial image, and this application does not specifically limit this.

[0107] In another example, the initial image is an image captured in vitro by medical personnel using the image acquisition end of an endoscope. Since medical personnel can directly determine whether the image is captured in vivo or in vitro when performing image capture, for example, based on the purpose of the capture, the image can be determined as a specimen image captured in vitro. That is, the medical personnel can manually identify whether the image is a first-category image, and by receiving the response information corresponding to the user's manual recognition, it can be quickly determined whether it is a first-category image. In this example, the above-mentioned step S2012 determines that the initial image is a first-category image based on the image information of the initial image, and can include the following method for determining the response information: in response to the operation information for the initial image, obtaining manual recognition information about the initial image; and obtaining response information for the initial image based on the manual recognition information.

[0108] Exemplarily, the operation information may be user input information (which may include text input or voice input, etc.), annotation information (for example, the user marks or annotates a specific area in the image to provide identification information about the area), gesture operation, etc.

[0109] Specifically, the manual identification information can be recorded in a variety of ways. For example, medical staff can directly input the manual identification information through an input device (such as a keyboard or touch screen) when collecting images, and obtain corresponding response information after the input is completed, such as Figure 7b Alternatively, the spoken judgment may be converted into text information through a speech recognition system. In addition, options may be provided through the user interface of the image acquisition device, such as Figure 7cAs shown, medical staff can select the image category during acquisition, thereby automatically generating corresponding identification information. In this way, manual identification information can be stored together with the image, facilitating subsequent classification and processing of the image. By way of further example, the above-mentioned step S2012 determines that the initial image is a first category image based on the image information of the initial image, and may include the following method of determining the image information: determining the target pixel of the initial image, wherein the target pixel is at least part of the pixels in the initial image after brightness screening; determining the hemoglobin index corresponding to the target pixel based on the channel values ​​of the red and blue channels corresponding to the target pixel; and determining the image information of the initial image based on the hemoglobin index corresponding to the target pixel.

[0110] Optionally, the above process of determining the target pixel of the initial image can be carried out as follows: for each pixel in the initial image, the brightness value corresponding to each pixel is determined based on the sum of the channel value of the red channel after each pixel is weighted by the first weight, the channel value of the green channel after each pixel is weighted by the second weight, and the channel value of the blue channel after each pixel is weighted by the third weight; the target pixel of the initial image is determined based on the brightness value of each pixel and the preset brightness value.

[0111] In this embodiment, considering that pixels with lower brightness have larger errors, if these pixels are combined for subsequent calculations, the results may be subject to certain errors. Therefore, in this embodiment, when counting the number of pixels, pixels with lower brightness are first excluded to determine the target pixel in the initial image. For example, the brightness of a certain pixel is calculated using the following formula:

[0112] Y=a*R+b*G+c*B

[0113] Wherein, Y represents the brightness of a pixel, R, G, and B are the values ​​of the red, green, and blue channels, respectively, and a, b, and c represent the first, second, and third weights, respectively. Those skilled in the art can make adaptive adjustments based on actual applications or prior data. For example, the values ​​can be 0.2129, 0.7148, and 0.0718. In this embodiment, by comparing Y with a preset brightness value (which can be adaptively determined by those skilled in the art based on actual applications or prior data, and the specific data is not limited in this embodiment), if Y is less than the preset brightness value, it means that the brightness is too low and is not included in the statistical range.

[0114] It's understandable that the red, green, and blue channels refer to the RGB color model in digital image processing. In this model, an image is represented using three basic color channels: red (R), green (G), and blue (B). Each channel is a grayscale image, representing the intensity of the corresponding color in the image. The red channel contains information about the intensity of the red component in the image. Each pixel value in the channel typically represents the intensity of red in that pixel, ranging from 0 to 255 (in an 8-bit image depth). The green channel contains information about the intensity of the green component in the image. Similar to the red channel, each pixel value represents the intensity of green. The blue channel contains information about the intensity of the blue component in the image. Each pixel value represents the intensity of blue.

[0115] In some embodiments, other methods may be used to filter out target pixels, or all pixels in the initial image may be directly used as target pixels.

[0116] Next, the determination of the image information of the initial image based on the hemoglobin index corresponding to the target pixel in the above step is further introduced. This can be achieved in the following manner: based on the hemoglobin index corresponding to the target pixel and a preset index threshold, the number of target pixels in the initial image whose hemoglobin index reaches the preset index threshold is determined; based on the number and the total number of target pixels in the initial image, a ratio is determined; and based on the ratio and the preset ratio threshold, the initial image is determined to be an image of the first category.

[0117] This embodiment takes into account that images captured in vivo and corresponding to in vitro specimens may exhibit different hemoglobin indices (IHB). For example, in an in vitro environment, tissue may lose some blood supply, resulting in a decrease in hemoglobin content. Based on this observation, this embodiment calculates the IHB index of target pixels in the image to determine image information of the initial image, and uses this image information to determine a first category image (i.e., the ratio of the number of target pixels to the total number of target pixels is less than a preset ratio threshold, which can be adaptively adjusted based on actual application or prior data).

[0118] Specifically, when in vivo, the vast majority of the endoscope's field of view is tissue mucosa, which appears red in the image. The hemoglobin index (IHB) can be used to determine whether it belongs to normal human tissue and, therefore, whether the image is taken in vivo. The IHB calculation formula is as follows:

[0119]

[0120] Where R is the value of the red channel of the image pixel, and G is the value of the green channel of the image pixel. When IHB is greater than a certain threshold (also known as a preset index threshold, which can be determined by those skilled in the art based on practical applications or experience), the pixel is considered to belong to human tissue; otherwise, it is considered to belong to non-tissue. If the proportion of pixels belonging to human tissue to the total number of pixels in the entire image is greater than the set threshold, the current image is considered to be in vivo; otherwise, it is an image captured of a specimen in vitro, i.e., the first category image.

[0121] In some embodiments, in addition to determining the image information of the endoscopic image by using the hemoglobin index,

[0122] Through the above method, it is possible to automatically identify whether the image currently taken by the endoscope is a normal image of human body tissue or an image taken in vitro of a specimen, and the identification result is relatively accurate.

[0123] In some embodiments, the above-mentioned step S203 performs distortion correction on the first category image based on a preset distortion correction algorithm to obtain a distortion-corrected image. The image correction processing can be performed by forward mapping. Specifically: according to the distortion correction algorithm, the pixel coordinates corresponding to each pixel of the first category image in the distortion-corrected image are determined; according to the pixel coordinates corresponding to each pixel of the first category image and the pixel value corresponding to the pixel coordinates, each pixel of the distortion-corrected image is assigned a value one by one to obtain the distortion-corrected image.

[0124] For example, a distortion correction algorithm can be used to calculate the coordinates (x', y') of each pixel coordinate (x, y) in the original image (i.e., the first category image with distortion) in the undistorted image, and convert the calculated ideal coordinates back to the pixel coordinate system and map them to the corresponding positions in the original image to obtain a distortion-corrected image.

[0125] Considering that during the forward mapping process, some pixel locations may not be covered by any original pixels, resulting in holes. To solve this problem, interpolation methods (such as bilinear interpolation) can be used to estimate the pixel values ​​of these hole locations. In addition, some pixel locations may be mapped to multiple original pixels. In this case, appropriate processing (such as averaging) is required to determine the final pixel value, thereby improving the accuracy of distortion correction.

[0126] By using the forward mapping method, starting from the first category image, the pixel values ​​are projected into the distortion-corrected image, making the distortion correction process more intuitive and efficient.

[0127] In other embodiments, the first category image may be subjected to distortion correction by inverse mapping. Specifically, step S203, based on a preset distortion correction algorithm, performs distortion correction on the first category image to obtain a distortion-corrected image. This may be performed in the following manner: determining the pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image according to the distortion correction algorithm; and assigning values ​​to each pixel of the distortion-corrected image one by one based on the pixel coordinates corresponding to each pixel of the distortion-corrected image and the pixel values ​​corresponding to the pixel coordinates to obtain the distortion-corrected image.

[0128] For example, the corresponding pixel coordinates in the original first-category image are calculated for each pixel in the distortion-corrected image based on the distortion correction algorithm. This process is equivalent to the inverse mapping of the distortion-corrected image, facilitating the subsequent extraction of information from the original image to construct the corrected image. For each pixel in the distortion-corrected image, the previously determined corresponding pixel coordinates are used to find the corresponding pixel value in the original first-category image, and the obtained pixel value is assigned to the corresponding pixel position in the distortion-corrected image. In this way, each pixel in the distortion-corrected image is assigned a value one by one. After all pixels have been assigned values, the resulting image is the distortion-corrected image.

[0129] Optionally, the distortion correction algorithm includes camera intrinsic parameters and distortion parameters. In the above steps, the pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image are determined according to the distortion correction algorithm. This can be done by: determining the image plane coordinates of each pixel of the distortion-corrected image based on the pixel coordinates of each pixel in the distortion-corrected image and the inverse matrix of the camera intrinsic parameters; and determining the pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image based on the image plane coordinates of each pixel in the distortion-corrected image and the distortion parameters.

[0130] For example, when performing distortion correction, this embodiment uses the distortion parameters of the algorithm and the camera intrinsic parameters. Assuming that the distorted image is I, the pixel coordinates are represented by (x, y, 0), and the undistorted image is II, the pixel coordinates are represented by (u, v, 0), the coordinates (u, v, 0) are converted to the undistorted image plane coordinates using the following formula (5).

[0131]

[0132] Then, substitute (x′, y′) into formula (3) and formula (4) to obtain the distorted coordinates (x, y, 0). That is, for each pixel II(u, v) in the distortion-corrected image, its value is equal to the pixel value I(x, y) in the distorted image.

[0133] The inverse mapping method described above can effectively correct image distortion, making the image more consistent with its actual geometry and proportions. Compared to forward mapping or other distortion correction methods, some pixels in the target image may not be covered by pixels in the original image, resulting in holes or undefined areas. Inverse mapping effectively avoids these holes by mapping the target image back to the original image pixel by pixel, thereby improving correction accuracy.

[0134] In some embodiments, after the distortion-corrected image is saved as a specimen image, the specimen image may be sent to a target terminal so that the target terminal stores or displays the specimen image; or displays the specimen image.

[0135] In one example, the target terminal may be Figure 1b The terminal device shown may also be another terminal that has the need to browse or process the specimen image, and this embodiment is not limited to this. For example, upon receiving a user-initiated request, the specimen image can be sent to the target terminal. User requests can be initiated in a variety of ways, such as through an application program interface, a web interface, or through voice commands. In some embodiments, the user can choose whether to send the specimen image immediately or under specific conditions (such as connecting to a specific network or at a specific time). In addition, after receiving the specimen image, the target terminal can perform various operations. For example, the target terminal can further process the specimen image, such as image enhancement, filtering, or analysis; alternatively, the target terminal can store the specimen image in a local storage device or upload it to a cloud storage service for backup and sharing. In some embodiments, a notification function can also be provided to notify the user after the specimen image is successfully sent or received, such as through a pop-up message, email, text message, etc. In this way, users can conveniently share and view specimen images between different devices, improving the flexibility and convenience of image processing and management.

[0136] In another example, by displaying the specimen image, users can immediately review the effects of the correction. If the user finds that the image quality does not meet expectations, the system can allow the user to make further adjustments or retake the image. This immediate feedback mechanism helps improve user satisfaction and image processing accuracy. In some embodiments, a multi-view display function can also be provided, that is, the original image and the corrected specimen image can be displayed simultaneously. This allows the user to intuitively compare the results before and after correction, helping them better understand the impact of the correction process.

[0137] In a second aspect, an embodiment of the present application provides a structural diagram of an image processing device for endoscopic images, such as Figure 8 As shown, the device 800 may include a first acquisition module 801, a correction module 802 and a storage module, wherein:

[0138] The first acquisition module 801 is used to acquire a first category of images; wherein the first category of images is used to indicate images acquired by an image acquisition end of an endoscope photographing a specimen in vitro;

[0139] A correction module 802 is configured to perform distortion correction on the first category image based on a preset distortion correction algorithm to obtain a distortion-corrected image, wherein the distortion correction algorithm is an algorithm for performing distortion correction on the first category image;

[0140] The saving module 803 is used to save the distortion-corrected image as a specimen image.

[0141] In one embodiment, the first acquisition module 801 includes:

[0142] An initial image acquisition unit, used to acquire an initial image acquired by the image acquisition end of the endoscope;

[0143] The category determination unit is configured to determine that the initial image is an image of the first category according to at least one of the image information, identification information, and response information of the initial image.

[0144] In one embodiment, the category determination unit includes: a pixel determination subunit, configured to determine a target pixel of an initial image, wherein the target pixel is at least a portion of pixels in the initial image after brightness filtering; an index determination subunit, configured to determine a hemoglobin index corresponding to the target pixel based on channel values ​​of red and blue channels corresponding to the target pixel; and a first determination subunit, configured to determine image information of the initial image based on the hemoglobin index corresponding to the target pixel;

[0145] And / or, a response subunit, configured to obtain manual recognition information about the initial image in response to operation information for the initial image; and a second determination subunit, configured to obtain response information of the initial image based on the manual recognition information.

[0146] In one embodiment, the information determination subunit is specifically used to: determine the number of target pixels in the initial image whose hemoglobin index reaches the preset index threshold based on the hemoglobin index corresponding to the target pixel and the preset index threshold; determine the ratio based on the number and the total number of target pixels in the initial image; and determine that the initial image is a first category image based on the ratio and the preset ratio threshold.

[0147] In one embodiment, the pixel determination subunit is specifically used to: for each pixel in the initial image, determine the brightness value corresponding to each pixel based on the sum of the channel value of the red channel of each pixel weighted by the first weight, the channel value of the green channel of each pixel weighted by the second weight, and the channel value of the blue channel of each pixel weighted by the third weight; determine the target pixel of the initial image based on the brightness value of each pixel and the preset brightness value.

[0148] In one embodiment, the correction module 802 includes:

[0149] A first coordinate determining unit, configured to determine, according to a distortion correction algorithm, pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category of images;

[0150] The first assignment unit is used to assign values ​​to each pixel of the distortion-corrected image one by one according to the pixel coordinates corresponding to each pixel of the distortion-corrected image and the pixel value corresponding to the pixel coordinates, so as to obtain the distortion-corrected image.

[0151] In one embodiment, the distortion correction algorithm includes camera intrinsic parameters and distortion parameters; the first coordinate determination unit is specifically used to: determine the image plane coordinates of each pixel of the distortion-corrected image based on the pixel coordinates of each pixel of the distortion-corrected image and the inverse matrix of the camera intrinsic parameters; determine the pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image based on the image plane coordinates of each pixel of the distortion-corrected image and the distortion parameters.

[0152] In one embodiment, the correction module 802 includes:

[0153] a second coordinate determining unit, for determining, according to a distortion correction algorithm, pixel coordinates corresponding to each pixel of the first category image in the distortion-corrected image;

[0154] The second assignment unit is used to assign a value to each pixel of the distortion-corrected image one by one according to the pixel coordinates corresponding to each pixel of the first category image and the pixel value corresponding to the pixel coordinates to obtain the distortion-corrected image.

[0155] In one embodiment, the device further comprises:

[0156] a sending module, configured to send the specimen image to a target terminal so that the target terminal stores or displays the specimen image; or

[0157] The display module is used to display the specimen image.

[0158] The relevant instructions can be understood by referring to the relevant descriptions and effects corresponding to the steps in the embodiment of the method of this application, and no further details will be given here.

[0159] Figure 9 An endoscope provided in an embodiment of the present application is as follows: Figure 9 As shown, the endoscope 900 includes: a processor 901, and a memory 902 communicatively connected to the processor 901;

[0160] The memory 902 stores computer-executable instructions;

[0161] The processor 901 executes the computer-executable instructions stored in the memory 902 to implement the specimen image generation method based on endoscope photography, wherein the memory 902 and the processor 901 are connected via a bus.

[0162] An embodiment of the present application also provides a computer-readable storage medium, which includes computer instructions. When the computer instructions are executed on a computer, the computer executes the specimen image generation method based on endoscopic photography provided by the above method embodiment.

[0163] Among them, the computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0164] The relevant instructions can be understood by referring to the relevant descriptions and effects corresponding to the steps in the embodiment of the method of this application, and no further details will be given here.

[0165] An embodiment of the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, it can implement the specimen image generation method based on endoscopic photography provided in the above method embodiment.

[0166] The relevant instructions can be understood by referring to the relevant descriptions and effects corresponding to the steps in the embodiment of the method of this application, and no further details will be given here.

[0167] It will be understood by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium).

[0168] As is well known to those skilled in the art, the term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0169] Furthermore, as is well known to those skilled in the art, communication media typically embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0170] In the description of the embodiments of the present application, the term "and / or" merely represents an association relationship that describes associated objects, indicating that three relationships may exist. For example, A and / or B can represent three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" represents any combination of at least two of any one or more of a plurality of items. For example, at least one of A, B, and C can represent any one or more elements selected from a set that includes A, B, and C. In addition, the term "plurality" means two or more, unless otherwise specified.

[0171] In the description of the embodiments of the present application, the terms "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0172] Finally, it should be noted that the above embodiments 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, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for generating a specimen image based on endoscope photography, characterized in that: include: Acquire a first category of images; wherein the first category of images is used to indicate images acquired by an image acquisition end of an endoscope photographing a specimen in vitro; Based on a preset distortion correction algorithm, performing distortion correction on the first category of images to obtain a distortion-corrected image, wherein the distortion correction algorithm is an algorithm for performing distortion correction on the first category of images; The distortion-corrected image is saved as a specimen image.

2. The method according to claim 1, characterized in that The acquiring of the first category image includes: Acquiring an initial image captured by an image acquisition terminal of the endoscope; According to at least one of the image information, identification information and response information of the initial image, it is determined that the initial image is an image of the first category.

3. The method according to claim 2, characterized in that The step of determining, based on the image information of the initial image, that the initial image is an image of the first category includes: Determining target pixels of the initial image, wherein the target pixels are at least a portion of pixels in the initial image after brightness filtering; Determining the hemoglobin index corresponding to the target pixel according to the channel values ​​of the red and blue channels corresponding to the target pixel; determining image information of the initial image according to the hemoglobin index corresponding to the target pixel; And / or, determining, based on the manual recognition information of the initial image, that the initial image is an image of the first category includes: In response to operation information on the initial image, acquiring human recognition information about the initial image; Response information of the initial image is obtained according to the manual recognition information.

4. The method according to claim 3, characterized in that The determining of the image information of the initial image according to the hemoglobin index corresponding to the target pixel includes: determining, in the initial image, the number of target pixels whose hemoglobin index reaches the preset index threshold according to the hemoglobin index corresponding to the target pixel and a preset index threshold; determining a ratio based on the number and the total number of target pixels in the initial image; According to the ratio and a preset ratio threshold, the initial image is determined to be an image of the first category.

5. The method according to claim 3 or 4, characterized in that The determining of the target pixel of the initial image includes: For each pixel in the initial image, determining a brightness value corresponding to each pixel according to the sum of a channel value of the red channel weighted by the first weight, a channel value of the green channel weighted by the second weight, and a channel value of the blue channel weighted by the third weight; The target pixel of the initial image is determined according to the brightness value of each pixel and a preset brightness value.

6. The method according to any one of claims 1 to 5, characterized in that The step of performing distortion correction on the first category image based on a preset distortion correction algorithm to obtain a distortion-corrected image includes: Determining pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image according to the distortion correction algorithm; According to the pixel coordinates corresponding to each pixel of the distortion-corrected image and the pixel values ​​corresponding to the pixel coordinates, each pixel of the distortion-corrected image is assigned a value one by one to obtain the distortion-corrected image.

7. The method according to claim 6, characterized in that The distortion correction algorithm includes camera intrinsic parameters and distortion parameters; and determining, according to the distortion correction algorithm, pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image, includes: Determining the image plane coordinates of each pixel of the distortion-corrected image according to the pixel coordinates of each pixel of the distortion-corrected image and the inverse matrix of the camera intrinsic parameter; According to the image plane coordinates of each pixel of the distortion-corrected image and the distortion parameter, pixel coordinates corresponding to each pixel of the distortion-corrected image in the first category image are determined.

8. The method according to any one of claims 1 to 5, characterized in that The step of performing distortion correction on the first category image based on a preset distortion correction algorithm to obtain a distortion-corrected image includes: determining, according to the distortion correction algorithm, pixel coordinates corresponding to each pixel of the first category image in the distortion-corrected image; According to the pixel coordinates corresponding to each pixel of the first category image and the pixel values ​​corresponding to the pixel coordinates, each pixel of the distortion-corrected image is assigned a value one by one to obtain the distortion-corrected image.

9. The method according to any one of claims 1 to 5, characterized in that After saving the distortion-corrected image as a specimen image, the method further includes: sending the specimen image to a target terminal so that the target terminal stores or displays the specimen image; or The specimen image is displayed.

10. An endoscope, characterized in that: include: memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the endoscope executes the method for generating a specimen image based on endoscopic photography according to any one of claims 1 to 9.

Images