Image display method, image display apparatus, and capsule endoscope system

Abstract

The present invention provides an image display method, an image display apparatus, and a capsule endoscope system. The method comprises: detecting a tissue site involved in detection images and a tissue area corresponding to the tissue site, on the basis of the tissue site involved in the detection images and the tissue area, performing calculation to obtain site position color information, and respectively displaying, on a scale axis, colors corresponding to the site position color information, to form site color bars. Site color bars are obtained by means of the method. The method is simple and easy to implement, achieves high accuracy and high speed, and allows for easy recognition of different tissue sites.

Description

Image display method, image display device and capsule endoscope system

[0001] Cross-references

[0002] This application claims priority to Chinese patent application number 202310993367.7, filing date August 9, 2023, and invention name “An image display method, image display device, and capsule endoscope system.” The entire contents of the above patent application are hereby incorporated into this application as a reference. Technical Field

[0003] The present invention relates to the field of medical equipment imaging, and in particular to an image display method, an image display device, and a capsule endoscope system. Background Art

[0004] Image display devices are widely used in the field of medical imaging. For example, a capsule endoscope is a medical device that can transmit captured images to a terminal for display. A capsule endoscope integrates core components such as a camera, wireless transmission antenna, and sensors into a capsule that can be swallowed by the human body. During the examination, the capsule endoscope typically uses various sensors to record various characteristics of the body. For example, a CMOS (Complementary Metal Oxide Semiconductor) sensor records digestive tract imaging characteristics, a pH (hydrogen ion concentration) sensor records the pH value of the digestive tract, and a temperature sensor records the temperature of the digestive tract. Among these characteristics recorded by the capsule endoscope, digestive tract imaging characteristics are the most important basis for doctors to make diagnoses.

[0005] Capsule endoscopes collect images of different tissue regions or parts within the body and simultaneously transmit them outside the body. Medical examinations are then performed based on the acquired image data. However, the large number of images presents a significant challenge for doctors in interpreting them. Conventional methods typically use the average color value of the image to generate a color scroll bar, which is then manipulated to retrieve and display the image.

[0006] There are two problems with the above method of generating a color scroll bar using the average color value of the image: first, the color scroll bar generated using the average color value does not take into account the difference in brightness between images, and there is no way to eliminate interference caused by food residues or other foreign matter in the human tissue area, resulting in a large deviation in the generated color scroll bar; second, the color scroll bar generated using the average color value cannot distinguish certain tissue parts, such as the small intestine and colon, which are both yellowish in color. The color bar generated using the average color makes it difficult to distinguish these two tissue parts, thereby affecting the retrieval and display of tissue parts during the film reading process, resulting in low accuracy, large errors, and a poor user experience.

[0007] Summary of the Invention

[0008] One of the objectives of the present invention is to provide an image display method to solve the technical problem in the prior art that a color scroll bar cannot accurately distinguish different tissue parts in a human tissue area.

[0009] In order to achieve one of the above-mentioned objects of the invention, the present invention provides an image display method, comprising: an input step of inputting detection images taken by a capsule endoscope in a time series; a scale axis display control step of controlling the scale axis to be displayed and the ruler that can move on the scale axis to be displayed; an image display control step of controlling the movement of the ruler on the scale axis to be linked with the switching of the image displayed on the image display unit, and displaying the detection image at the shooting moment corresponding to the position of the ruler on the image display unit; a part position information detection step of inputting the detection images input by the input step into a part position image recognition model for tissue part information detection, wherein the tissue part information includes a first tissue part to which the detection image belongs, a first tissue part to which the detection image belongs, and a second tissue part to which the detection image belongs. a first probability value p of a first tissue site and a first tissue region to which the first tissue site belongs; a site position color information detection step, which calculates the site position color information of the detection image based on the tissue site information of the detection image detected by the site position information detection step; a site position color display control step, which controls so that the colors corresponding to the site position color information of the detection image calculated by the site position color information detection step are respectively displayed at positions corresponding to time on the scale axis to form site color bars; a segmented position information detection step, which inputs the first tissue site to which the detection image belongs output by the site position information detection step into a segmented position image recognition model to perform segmented position information detection, wherein the segmented position information includes the segmented region to which the detection image belongs;

[0010] A segmented position color information detection step is performed to set the segmented position color information corresponding to the segmented area to which the detection image belongs according to the segmented area to which the detection image belongs outputted from the segmented position information detection step; a segmented position color display control step is performed to control so that the colors corresponding to the segmented position color information corresponding to the segmented area to which the detection image belongs set by the segmented position color information detection step are respectively displayed at positions corresponding to time on the scale axis to form segmented color bars.

[0011] In order to achieve one of the above-mentioned objects of the invention, the present invention provides a capsule endoscope system, comprising: a capsule endoscope, a data receiving terminal, and the image display device as described above.

[0012] Compared with the prior art, the embodiments of the present invention have at least one of the following beneficial effects:

[0013] The image display method adopted by the present invention detects the tissue part and the corresponding tissue area where the shooting position of the detection image is located in the human body in the part position information detection step, and then calculates the part position color information corresponding to the detection image according to the tissue part and the corresponding tissue area detected in the part position information detection step in the part position color information detection step. Finally, the color corresponding to the part position color information is displayed at the position corresponding to the time on the scale axis to form a part color bar, thereby realizing the display of the tissue part where the shooting position of the detection image is located in the human body in the form of a part color bar, so that the doctor can directly know the tissue part where the shooting position of the detection image is located in the human body by identifying the color on the part color bar. The part color bar in the present invention is simple and easy to implement, and has high accuracy, fast speed and high efficiency in retrieving and checking detection images, and is easy to identify different tissue parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] FIG. 1 is a schematic diagram showing steps of an image display method according to an embodiment of the present invention.

[0015] FIG. 2 is a schematic diagram showing specific steps of step S4 of the image display method according to one embodiment of the present invention.

[0016] FIG. 3 is a partial schematic diagram of a first embodiment of step S5 of an image display method according to an embodiment of the present invention.

[0017] FIG. 4 is a partial schematic diagram of a second embodiment of step S5 of the image display method according to one embodiment of the present invention.

[0018] FIG. 5 is a partial schematic diagram of a third embodiment of step S5 of the image display method according to an embodiment of the present invention.

[0019] FIG. 6 is a schematic diagram showing specific steps of step S7 of the image display method according to one embodiment of the present invention.

[0020] FIG. 7 is a schematic structural diagram of a capsule endoscope system according to an embodiment of the present invention.

[0021] FIG8 is a schematic structural diagram of an image display device according to an embodiment of the present invention.

[0022] FIG9 is a schematic structural diagram of a playback operation unit in one embodiment of the present invention.

[0023] FIG. 10 is a schematic structural diagram of a color scroll bar unit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be described in detail below with reference to the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional changes made by those skilled in the art based on these embodiments are all within the scope of protection of the present invention.

[0025] It should be noted that the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In addition, the terms "first," "second," "third," etc., are used for descriptive purposes only and are not to be understood as indicating or implying relative importance.

[0026] In the field of medical device imaging, the image characteristics of different tissue areas of the human body are the most important basis for doctors to make diagnoses. Usually, these image characteristics can be obtained by time-sequential photography through a capsule endoscope, and the number of images captured at a time may reach tens of thousands. In order to facilitate doctors to quickly and efficiently retrieve and review images, the large number of images captured needs to be processed.

[0027] Based on this, an embodiment of the present invention provides an image display method, as shown in FIG1 , the method may include:

[0028] Step S1, an input step, inputting detection images taken by a capsule endoscope in a time series;

[0029] Step S2, a scale axis display control step, performing control to display a scale axis and a ruler movable on the scale axis, wherein the scale axis represents the entire shooting period of the detection image captured in time series input in the aforementioned input step;

[0030] Step S3, an image display control step, performing control so that the movement of the ruler on the scale axis is linked to the switching of the image displayed on the image display unit, and the detection image at the shooting time corresponding to the position of the ruler is displayed on the image display unit;

[0031] Step S4, a part position information detection step, inputting the detection images input in the input step into a part position image recognition model to perform tissue part information detection, wherein the tissue part information includes a first tissue part to which the detection image belongs, a first probability value p of the detection image belonging to the first tissue part, and a first tissue region to which the first tissue part belongs;

[0032] Step S5, a part position color information detection step, calculating the part position color information of the detection image based on the tissue part information of the detection image detected in the part position information detection step;

[0033] Step S6, a part position color display control step, performing control so that the colors corresponding to the part position color information of the detection image calculated in the part position color information detection step are displayed at positions corresponding to time on the scale axis, forming part color bars;

[0034] Step S7, a segmented position information detection step, inputting the first tissue part to which the detection image belongs outputted from the part position information detection step into a segmented position image recognition model to perform segmented position information detection, wherein the segmented position information includes the segmented area to which the detection image belongs;

[0035] Step S8, a segment position color information detection step, setting the segment position color information corresponding to the segment area to which the detection image belongs according to the segment area to which the detection image belongs outputted in the segment position information detection step;

[0036] Step S9, the segmented position color display control step, performs control so that the colors corresponding to the segmented position color information corresponding to the segmented area to which the detection image set by the segmented position color information detection step belongs are displayed at the positions corresponding to the time on the scale axis, respectively, to form a segmented color bar.

[0037] In this way, by controlling the linkage between the site color bar and / or segmented color bar and the detection image, different tissue sites and / or segmented areas can be quickly and accurately located, which is simple and easy to implement.

[0038] Furthermore, the part color bar is formed above the scale axis, and the segment color bar is formed below the scale axis.

[0039] As shown in FIG2 , in one embodiment, the part position information detection step (corresponding to step S4 ) may specifically include the following steps:

[0040] Step S41, detecting and obtaining a first tissue site to which the detection image belongs and a first probability value p of the detection image belonging to the first tissue site;

[0041] Step S42 , obtaining the first tissue region to which the detection image belongs according to the first tissue site to which the detection image belongs.

[0042] In this way, the calculation process is simple and easy to implement. By analyzing the recognition model, the tissue part to which the detection image belongs is determined according to the probability value, and the corresponding detection image can be retrieved quickly and intuitively. The operation is simple, making it convenient for doctors to retrieve and check at any time.

[0043] The first tissue region and the first tissue site do not specifically refer to a certain tissue region and a certain tissue site of the human tissue structure, and the first tissue region may also refer to an extracorporeal region. Specifically, the first tissue region is one of the extracorporeal region, the esophageal region, the gastric region, the small intestine region, and the large intestine region. When the first tissue region is the extracorporeal region, the first tissue site is one of the extracorporeal region and the oral cavity; when the first tissue region is the esophageal region, the first tissue site is one of the esophagus and the dentate line; when the first tissue region is the gastric region, the first tissue site is one of the fundus, the cardia, the anterior and posterior wall of the gastric body, the lesser curvature of the stomach, the greater curvature of the stomach, the gastric angle, the anterior and posterior wall of the gastric angle, the gastric antrum, the lesser curvature of the gastric antrum, the pylorus, and the unopened gastric antrum; when the first tissue region is the small intestine region, the first tissue site is one of the duodenal bulb, the descending duodenum, the small intestine, the jejunum, and the ileum; when the first tissue region is the large intestine region, the first tissue site is one of the ileocecal valve, the appendix opening, the colon, the rectum, and the anal canal.

[0044] In summary, there are five tissue regions, each of which has a corresponding ID identifier i. The ID identifier i can be used to uniquely identify the corresponding tissue region, where the value range of i is [1, 5]. For example, when i = 1, the corresponding tissue region is an in vitro region. There are 25 tissue sites, each of which has a corresponding ID identifier k. The ID identifier k can be used to uniquely identify the corresponding tissue site, where the value range of k is [0, 24]. For example, when k = 0, the corresponding tissue site is in vitro.

[0045] In this way, the regional and part division of human tissue structure can facilitate doctors to quickly locate the tissue area in advance and shorten the retrieval time; and then further accurately locate the tissue part to ensure that users take into account both timeliness and accuracy when performing image retrieval and review, which is conducive to improving work efficiency.

[0046] The step S41 may specifically include: inputting the detection image input in the input step into the part position image recognition model, and calculating a number of probability values ​​that the detection image belongs to any tissue part; traversing and screening the several probability values, taking the highest probability value as the first probability value p, and taking the tissue part corresponding to the first probability value p as the first tissue part to which the detection image belongs.

[0047] The first probability value p is in the range of (1 / N, 1], where N is the number of all tissue parts in all tissue regions of the human body. Preferably, when the tissue parts are divided into 25, that is, N is 25, the first probability value p is in the range of (1 / 25, 1].

[0048] In this way, in step S41, several probability values ​​of the detection image belonging to any tissue part are calculated through the part position image recognition model, and the tissue part corresponding to the highest probability value among the several probability values ​​is used as the first tissue part of the detection image. The tissue part corresponding to the shooting position of the detection image determined in this way has high accuracy and reliability.

[0049] The step S42 may specifically include:

[0050] The first tissue site to which the detected image belongs is input into a mapping table of human tissue sites and tissue regions to obtain the first tissue region to which the detected image belongs. For example, when the first tissue site is detected as a gastric angle, the first tissue site is input into the mapping table of human tissue sites and tissue regions to obtain the first tissue region as a stomach region.

[0051] Regarding step S5, the color information of the part position of the detected image includes a brightness value V, a hue value H, and a saturation value S.

[0052] As shown in FIG3 , in the first embodiment, the step of calculating the brightness value V corresponding to the detection image includes:

[0053] Step S511, determining an initial brightness value V' corresponding to the detection image according to the first tissue region to which the detection image belongs;

[0054] Step S512: Calculate an additional brightness value V" corresponding to the detection image based on the brightness distribution coefficient v corresponding to the detection image and the first probability value p, where the additional brightness value V" corresponding to the detection image = p*v.

[0055] Step S513 , calculating the brightness value V corresponding to the detection image according to the initial brightness value V′ and the additional brightness value V″ corresponding to the detection image, wherein the brightness value V corresponding to the detection image=V′+V″.

[0056] As shown in FIG4 , in the second embodiment, the step of calculating the hue value H corresponding to the detected image includes:

[0057] Step S521 , determining an initial hue value H′ corresponding to the detection image according to the first tissue region to which the detection image belongs;

[0058] Step S522 : Calculate the hue value H corresponding to the detection image according to the initial hue value H′ corresponding to the detection image, where the hue value H corresponding to the detection image=H′.

[0059] As shown in FIG5 , in the third embodiment, the step of calculating the saturation value S corresponding to the detection image includes:

[0060] Step S531 , determining an initial saturation value S′ corresponding to the detection image according to the first tissue region to which the detection image belongs;

[0061] Step S532, determining a position coefficient x corresponding to the detection image according to the first tissue region to which the detection image belongs;

[0062] Step S533, determining a saturation distribution value q corresponding to the detection image according to the first tissue site to which the detection image belongs;

[0063] Step S534: Calculate the additional saturation value S" corresponding to the detection image according to the position coefficient x and the saturation distribution value q corresponding to the detection image, where the additional saturation value S" corresponding to the detection image is equal to x*q.

[0064] Step S535 , calculating the saturation value S corresponding to the detection image based on the initial saturation value S′ and the additional saturation value S″ corresponding to the detection image, wherein the saturation value S corresponding to the detection image=S′+S″.

[0065] Correspondingly, the calculation process of the color information of the position of the detected image in step S5 can be expressed by the following formulas (1), (2) and (3):

[0066] V=V′+p*v (1)

[0067] H=H′ (2)

[0068] S=S′+x*q (3)

[0069] Among them, the initial brightness value V′, initial hue value H′, initial saturation value S′, and position coefficient x corresponding to all detection images taken in the same tissue area are the same; the brightness distribution coefficient v corresponding to all detection images is the same; and the saturation distribution value q corresponding to all detection images taken in the same tissue part is the same.

[0070] In order to facilitate doctors to distinguish and identify the colors corresponding to the various tissue parts on the part color bar formed in step S6, the following are the value settings of each parameter in a preferred embodiment:

[0071] The value range of the ID mark i of the tissue region is [1, 5], the value range of the ID mark k of the tissue part is [0, 24], and the brightness distribution coefficient v is 1 / 10;

[0072] When i=1, that is, the tissue region is an in vitro region, k is 0 or 1, V′ is set to 0.5, H′ and S′ are set to 0, x is set to 0.02, and the saturation distribution value q is set to k, that is,

[0073] V=0.5+p / 10; H=0; S=0.02k;

[0074] When i=2, that is, the tissue region is the esophagus region, k is 2 or 3, V′ is set to 0.75, H′ is set to 15, S′ is set to 0.35, x is set to 0.1, and the saturation distribution value q is set to (k-2), that is,

[0075] V=0.75+p / 10; H=15; S=0.35+0.1*(k-2);

[0076] When i=3, that is, the tissue region is the stomach region, k is set to 4 to 14, V′ is set to 0.65, H′ is set to 5, S′ is set to 0.26, x is set to 0.02, and the saturation distribution value q is set to (k-4), that is,

[0077] V=0.65+p / 10; H=5; S=0.26+0.02*(k-4);

[0078] When i=4, that is, the tissue region is the small intestine region, k is set to 15 to 19, V′ is set to 0.75, H′ is set to 50, S′ is set to 0.35, x is set to 0.03, and the saturation distribution value q is set to (k-15), that is,

[0079] V=0.75+p / 10; H=50; S=0.35+0.03*(k-15);

[0080] When i=5, that is, the tissue region is the large intestine region, k is set to 20 to 24, V′ is set to 0.55, H′ is set to 30, S′ is set to 0.6, x is set to 0.02, and the saturation distribution value q is set to (k-20), that is,

[0081] V=0.55+p / 10; H=30; S=0.6+0.02*(k-20).

[0082] In this way, since the initial hue values ​​H′ set for the detection images in different tissue areas are different, the hue values ​​H calculated for the detection images in different tissue areas are different. At the same time, since the saturation distribution values ​​q set for the detection images in different tissue parts within the same tissue area are different, the saturation values ​​S calculated for the detection images in different tissue parts within the same tissue area are different, which makes it convenient for doctors to distinguish and identify the colors corresponding to the detection images in different tissue parts within the same tissue area on the part color bar formed in step S6. Therefore, the part color bar formed in step S6 is intuitive, highly recognizable, has small errors, and is highly efficient.

[0083] For step S6, the part position color display control step is controlled so that the colors corresponding to the part position color information of the detection image calculated by the part position color information detection step are displayed at the positions corresponding to the time on the scale axis to form a part color bar.

[0084] Thus, first, in step S4, the tissue part and the corresponding tissue area where the shooting position of the detection image is located in the human body are detected, and then in step S5, the part position color information (including brightness value V, hue value H and saturation value S) corresponding to the detection image is calculated based on the tissue part and the corresponding tissue area detected in step S4, and finally the color corresponding to the part position color information is displayed at the position corresponding to the time on the scale axis to form a part color bar, thereby realizing that the tissue part where the shooting position of the detection image is located in the human body is displayed in the form of a part color bar, so that the doctor can directly know the tissue part where the shooting position of the detection image is located in the human body by identifying the color on the part color bar. The part color bar in the present invention is simple and easy to implement, and has high accuracy, fast speed and high efficiency in retrieving and consulting the detection image, and is easy to identify different tissue parts.

[0085] It should be noted that the same tissue region may correspond to multiple different tissue parts. In order to accurately and quickly locate the detection image corresponding to a certain tissue part on the scale axis, it is preferred to first locate the segmented area corresponding to the tissue part on the scale axis.

[0086] To this end, the image display method also includes step S7, a segmented position information detection step, in which the first tissue part to which the detection image output by the part position information detection step belongs is input into a segmented position image recognition model for segmented position information detection, and the segmented position information includes the segmented area to which the detection image belongs.

[0087] As shown in FIG6 , the segment position information detection step (step S7) may specifically include:

[0088] Step S71, obtaining the first tissue part to which the detection image outputted from the part position information detection step belongs;

[0089] Step S72, calculating the segmentation critical points of the first tissue part using a segmentation position image recognition model according to the first tissue part;

[0090] Step S73 : Segmenting the first tissue part to which the detection image belongs according to the segmentation critical point to obtain the segmented area to which the detection image belongs.

[0091] Specifically, the detection image input in the input step is input into the part position image recognition model for image recognition. Assuming that the first tissue part recognition result is {A,A,…,A,B,B,…,B,C,…,C}, the first tissue part recognition result is then input into the segmented position image recognition model. The segmented position image recognition model analyzes its position distribution through statistical distribution laws, determines that the dividing points of the segmented areas are the positions where A, B and C appear for the last time, and segments the first tissue part to obtain the segmented areas to which the detection image belongs.

[0092] In this way, demarcation and segmentation are performed based on the recognition result of the first tissue part to obtain segmented areas, which are convenient for doctors to view and analyze.

[0093] Furthermore, segmented color bars corresponding to different segmented areas may be constructed through the segmented position color detection step (corresponding to step S8 ) and the segmented position color display control step (corresponding to step S9 ).

[0094] Specifically, the step S8 includes: a segment position color information detection step, and setting the segment position color information corresponding to the segment area to which the detection image belongs according to the segment area to which the detection image belongs outputted from the segment position information detection step.

[0095] The step S9 includes: a segmented position color display control step, which controls so that the colors corresponding to the segmented position color information corresponding to the segmented area to which the detection image belongs, set by the segmented position color information detection step, are displayed respectively at positions corresponding to the time on the scale axis to form segmented color bars.

[0096] Preferably, the segmented position color information corresponding to the segmented area is pre-set, that is, the color and color order on the segmented color bar are fixed. The only difference is that the positions of the dividing points between the colors on the segmented color bar displayed for capsule endoscopy examinations on different subjects may be different.

[0097] Preferably, the pre-set segmented position color information corresponding to the segmented area can be used as the initial brightness value V', initial hue value H', and initial saturation value S' of the first tissue area. For example, the color of the stomach area on the segmented color bar can be pre-set to blue, and the brightness value, hue value, and saturation value of the blue can be used as the initial brightness value V', initial hue value H', and initial saturation value S' of the first tissue area. The calculated color of the part color bar corresponding to the gastric fundus, cardia, and other gastric tissue parts is light blue, dark blue, or other colors close to blue. In this way, the colors corresponding to the part color bar and the segmented color bar at the same position on the scale axis are similar, so that doctors can compare and analyze the part color bar and the segmented color bar when reading the film.

[0098] Therefore, when a doctor wants to quickly locate the detection image within tissue site A while reading an image, he can first locate the segmented area B corresponding to the tissue site A according to the different colors on the segmented color bar, and then locate the tissue site A within the scale axis range corresponding to the segmented area B according to the different colors on the site color bar, and finally move the ruler on the scale axis to the scale axis range corresponding to the tissue site A to display the detection image within the tissue site A.

[0099] Regarding the first tissue region and the segmented region, it should be noted that during the tissue part identification process of the test image, if the results of the part position image recognition model are correct, then the first tissue region to which a test image belongs according to the part position information detection step (i.e., step S4) and the segmented region to which the test image belongs according to the segmented position information detection step (i.e., step S7) can be understood as pointing to the same region. However, when the results of the part position image recognition model are incorrect, the first tissue region and the segmented region corresponding to the test image may point to different regions.

[0100] For example, suppose a certain detection image corresponds to the tissue part of the descending duodenum, but after detection in the part position information detection step (i.e., step S4), it is mistakenly identified as the fundus of the stomach. At this time, according to the mapping relationship, it can be known that the first tissue area corresponding to the tissue part (fundus of the stomach) is the stomach. However, in the segmented position information detection step (i.e., step S7), statistical calculations can be performed based on the first tissue part identification result, and it can be obtained that the tissue part should be located in the segmented area of ​​the small intestine. At this time, the first tissue area and the segmented area point to different areas. The first tissue area is an erroneous detection result, and the segmented area is a correct detection result. The part position detection result of the detection image is finally displayed in the form of a part color bar. Since the misidentification of the part position image recognition model is accidental, a small number of part position misidentifications of the part position image recognition model will not have a significant impact on the formation of the part color bar, will not affect the doctor's operation and use, and does not need to be corrected.

[0101] The segmentation position detection result of detection image is finally shown in the form of segmentation color bar, owing to needing to know the accurate dividing point between each segmentation area corresponding to detection image just can obtain accurate segmentation color bar, the segmentation position misidentification result of a detection image will cause significant impact on the division of segmentation area (i.e., the formation of segmentation color bar is caused significant impact).If directly adopt the first tissue area corresponding to the first tissue position corresponding to the detection image obtained by position position image recognition model to carry out segmentation area boundary division, the misidentification result that position position image recognition model occurs can cause significant error to occur in the boundary division of segmentation area, thereby obtain wrong segmentation color bar.Therefore, it is necessary to set the segmentation position information detection step (i.e. step S7) to find the dividing point of accurate segmentation area, to correct the first tissue area misidentification in the position position information detection step (i.e. step S4), thereby detect and obtain correct segmentation area boundary division by the segmentation position information detection step (i.e. step S7), finally obtain correct segmentation color bar.

[0102] In this way, the present invention can correct the misidentification that occurs in the part position information detection step through the segmented position information detection step, so that the segmented color bar obtained through the segmented position color information detection step and the segmented position color display control step is simple and easy to implement, and the retrieval and review of the detection image is high in accuracy, speed and efficiency, and different segmented areas can be easily identified.

[0103] When searching or reviewing detection images pointing to different tissue parts, it is also possible to further determine whether the detection images contain lesions. Specifically, the image display method may further include:

[0104] Step S10 is a lesion information detection step, in which the detection images inputted in the input step are respectively inputted into an image lesion recognition model to detect lesion information, wherein the lesion information includes whether a lesion area exists in the detection image. Specifically, step S10 may include the following steps:

[0105] Step S101, inputting the detection images input in the input step into an image lesion recognition model to obtain a second probability value of whether a lesion area exists in the detection image;

[0106] Step S102, comparing whether the second probability value of the detection image is within a preset probability range;

[0107] If so, the process jumps to step S103 to detect whether there is a lesion area in the detection image.

[0108] Specifically, the preset probability range is [0.5, 1], that is, when the second probability value of the detection image is between (0, 0.5), the detection image is detected as a normal image and there is no lesion area; when the second probability value is between [0.5, 1], the detection image is detected as an abnormal image and there is a lesion area.

[0109] Furthermore, the image display method may also include step S11, a marking point display control step, which controls the detection image of the lesion area detected by the lesion information detection step to display marking points at positions corresponding to time on the scale axis.

[0110] Furthermore, the marking point is formed below the segmented color bar.

[0111] In this way, the lesion information of the detection image can be displayed in the form of marked points below the scale axis. When the doctor is reading the image, he moves the ruler on the scale axis to the scale axis position corresponding to the marked point, and can quickly locate and check the detection image of the lesion area, so as to make a timely and effective diagnosis.

[0112] Furthermore, the image display method further includes step S12, a playback operation control step, in which operation control is performed to adjust the reading mode (such as automatic playback, skipping to the first / last image), playback speed, etc. of displaying the detection image.

[0113] Furthermore, the image display method further includes step S13, a 3D position display control step, which controls to display the camera position (in-vivo position) of the detection image displayed in the image display control step in 3D (three-dimensional).

[0114] Furthermore, the image display method further includes step S14, a summary display control step, in which control is performed to display in the summary display unit a detection image including a detection image pointing to a special segmented area, a special tissue site, or an area with a lesion. The detection image displayed in the summary display unit is selected, and the image display unit in the image display control step can be controlled to quickly associate and output a display corresponding to the selected detection image.

[0115] So far, the image display method provided in this embodiment has been described. Now, the image display device provided by the present invention will be described.

[0116] It should be noted that the image display device can be implemented independently, and the image display device can also correspond to the image display method. The corresponding descriptions and configurations can be implemented with each other and will not be repeated here.

[0117] As shown in FIG7 , the present invention further provides a capsule endoscope system, comprising: a capsule endoscope 1, a data receiving terminal 2, and an image display device 3. The capsule endoscope 1 can compress collected image data and send the compressed data to the data receiving terminal 2 via wireless communication, and the data receiving terminal 2 can also send the received image data to the image display device 3 via wireless communication or wired communication.

[0118] Specifically, the subject orally swallows the capsule endoscope 1, which then passes through the esophagus and advances within the body cavity, utilizing peristalsis of the digestive tract or controlled by an external control device. Capsule endoscope 1 sequentially captures detection images of the body cavity in a time series. These detection images are input to an image display device 3 via an input unit. Image display device 3 decodes and recognizes the image data, determining at least the tissue region, tissue location, and lesion corresponding to the image data, and then outputs the capsule endoscope image via an image display control unit.

[0119] Specifically, as shown in FIG8 , the image display device 3 includes:

[0120] An input unit (not shown in FIG8 ) that inputs detection images captured by the capsule endoscope in a time series.

[0121] The image display unit 10 is used to display the detection image.

[0122] The play operation unit 20 is used to adjust the reading mode (such as automatic play, jump to the first / last image), play speed, etc. of the capsule endoscope image.

[0123] As shown in FIG9 , the play operation unit 20 may specifically include: a play speed adjustment 201, a first image jump button 202, a forward play group of images button 203, a forward automatic play button 204, a backward automatic play button 205, a backward play group of images button 206, and a last image jump button 207. The forward or backward buttons are used to output and display the group of capsule endoscopy images preceding or following the currently played capsule endoscopy image according to the arrangement order.

[0124] The color scroll bar unit 30 is used to display the color corresponding to the tissue part and segmented area where the detection image is shot in the human body, and to mark the detection image where the lesion area exists.

[0125] As shown in FIG10 , the color scroll bar unit 30 may specifically include: a ruler 301 , a part color bar 302 , a segment color bar 303 , a marking point 304 , and a scale axis 305 (time axis).

[0126] The scale axis 305 (time axis) represents the entire capture period of the detection images captured in a time series input by the input unit. The scale 301 is movable on the scale axis 305. The movement of the scale 301 on the scale axis 305 is synchronized with the switching of the detection images displayed on the image display unit 10. The detection image corresponding to the capture time corresponding to the position of the scale 301 is displayed on the image display unit 10.

[0127] The part color bar 302 is used to display the color corresponding to the tissue part where the shooting position of the detection image in the human body is located.

[0128] The segmented color bar 303 is used to display the color corresponding to the segmented area where the shooting position of the detection image in the human body is located.

[0129] The marking point 304 is used to mark whether there is a lesion area in the detection image.

[0130] The 3D position display unit 40 is configured to display the imaging position (in-vivo position) of the detection image displayed on the image display unit 10 in 3D (three dimensions).

[0131] The summary display unit 50 is used to display detection images including those pointing to special segmented areas, special tissue sites, or areas with lesions. By selecting an image in the summary display unit 50, the image display unit 10 can be controlled to quickly associate and output the corresponding selected detection image, making it convenient for doctors to quickly view tissue sites and lesion areas.

[0132] The detection unit (not shown in FIG8 ) includes a part position detection unit and a segment position detection unit.

[0133] The part position detection unit includes a part position information detection unit and a part position color information detection unit.

[0134] The part position information detection unit inputs the detection images input by the input unit into the part position image recognition model to perform tissue part information detection, where the tissue part information includes the first tissue part to which the detection image belongs, the first probability value p of the detection image belonging to the first tissue part, and the first tissue area corresponding to the first tissue part.

[0135] The part position color information detection unit calculates the part position color information of the detection image based on the tissue part information of the detection image detected by the part position information detection unit.

[0136] The segment position detection unit includes a segment position information detection unit and a segment position color information detection unit.

[0137] The segmented position information detection unit inputs the first tissue part to which the detection image output by the part position information detection unit belongs into the segmented position image recognition model to perform segmented position information detection, wherein the segmented position information includes the segmented area to which the detection image belongs.

[0138] The segment position color information detection unit sets the segment position color information corresponding to the segment area to which the detection image belongs according to the segment area to which the detection image belongs output by the segment position information detection unit.

[0139] The control unit (not shown in FIG8 ) includes a scale display control unit, an image display control unit, a part position color display control unit, and a segment position color display control unit.

[0140] The scale axis display control unit controls to display the scale axis 305 and the scale 301 that can move on the scale axis 305, wherein the scale axis 305 represents the entire shooting period of the detection image captured in time series input by the input unit.

[0141] The image display control unit controls the movement of the scale 301 on the scale axis 305 and the switching of the detection image displayed in the image display unit 10 so as to synchronize the detection image at the shooting moment corresponding to the position of the scale 301 on the image display unit 10.

[0142] The part position color display control unit controls so that the colors corresponding to the part position color information of the detection image calculated by the part position color information detection unit are displayed at positions corresponding to time on the scale axis, forming a part color bar 302.

[0143] The segmented position color display control unit controls so that the colors corresponding to the segmented position color information corresponding to the segmented area to which the detection image belongs, set by the segmented position color information detection unit, are displayed at positions corresponding to the time on the scale axis, forming a segmented color bar 303.

[0144] Preferably, the part color bar 302 is formed above the scale axis 305 , and the segment color bar 303 is formed below the scale axis 305 .

[0145] Specifically, the part color bar 302 and the segmented color bar 303 are formed at corresponding positions of the scale axis 305. By controlling the movement of the scale 301 on the scale axis 305 and synchronously switching the detection image displayed in the image display unit 10, the detection image at the shooting moment corresponding to the position of the scale 301 is displayed on the image display unit 10, which facilitates fast and accurate image retrieval with high efficiency and strong flexibility.

[0146] Furthermore, the part position information detection unit further includes a first tissue part detection unit and a first tissue region detection unit.

[0147] Regarding the first tissue site detection unit, the first tissue site detection unit can input the detection image input by the input unit into the site position image recognition model, and calculate several probability values ​​that the detection image belongs to any tissue site; the first tissue site detection unit traverses and screens the several probability values, takes the highest probability value as the first probability value p, and takes the tissue site corresponding to the first probability value p as the first tissue site to which the detection image belongs.

[0148] Regarding the first tissue area detection unit, the first tissue area detection unit can input the first tissue part to which the detection image belongs detected by the first tissue part detection unit into a mapping table of human tissue part-tissue area to obtain the first tissue area to which the detection image belongs.

[0149] In this way, by identifying and analyzing the detection image, the tissue part corresponding to the detection image is determined by the highest probability value. The method is simple, easy to implement, and has a high recognition accuracy.

[0150] Furthermore, the part position color information of the detected image includes a brightness value V, a hue value H and a saturation value S, and the part position color information detection unit includes a brightness value detection unit, a hue value detection unit and a saturation value detection unit.

[0151] Regarding the brightness value detection unit, the brightness value detection unit determines the initial brightness value V′ corresponding to the detection image based on the first tissue area to which the detection image belongs detected by the first tissue area detection unit; the brightness value detection unit calculates the additional brightness value V" corresponding to the detection image based on the brightness distribution coefficient v corresponding to the detection image and the first probability value p of the detection image detected by the first tissue part detection unit, and the additional brightness value V" corresponding to the detection image = p*v; the brightness value detection unit calculates the brightness value V corresponding to the detection image based on the initial brightness value V′ and the additional brightness value V" corresponding to the detection image, and the brightness value V corresponding to the detection image = V′+V".

[0152] Regarding the hue value detection unit, the hue value detection unit determines the initial hue value H′ corresponding to the detection image based on the first tissue area to which the detection image belongs detected by the first tissue area detection unit; the hue value detection unit calculates the hue value H corresponding to the detection image based on the initial hue value H′ corresponding to the detection image, and the hue value H corresponding to the detection image = H′.

[0153] Regarding the saturation value detection unit, the saturation value detection unit determines the initial saturation value S′ corresponding to the detection image based on the first tissue area to which the detection image belongs detected by the first tissue area detection unit; the saturation value detection unit determines the position coefficient x corresponding to the detection image based on the first tissue area to which the detection image belongs detected by the first tissue area detection unit; the saturation value detection unit determines the saturation distribution value q corresponding to the detection image based on the first tissue part to which the detection image belongs detected by the first tissue part detection unit; the saturation value detection unit calculates the additional saturation value S" corresponding to the detection image based on the position coefficient x and the saturation distribution value q corresponding to the detection image, and the additional saturation value S" corresponding to the detection image = x*q; the saturation value detection unit calculates the saturation value S corresponding to the detection image based on the initial saturation value S′ and the additional saturation value S" corresponding to the detection image, and the saturation value S corresponding to the detection image = S′+S".

[0154] Correspondingly, the calculation process of the part position color information of the detection image in the part position color information detection unit can be expressed by the following formulas (1), (2) and (3):

[0155] V=V′+p*v (1)

[0156] H=H′ (2)

[0157] S=S′+x*q (3)

[0158] Among them, the initial brightness value V′, initial hue value H′, initial saturation value S′, and position coefficient x corresponding to all detection images taken in the same tissue area are the same; the brightness distribution coefficient v corresponding to all detection images is the same; and the saturation distribution value q corresponding to all detection images taken in the same tissue part is the same.

[0159] In order to facilitate doctors to distinguish and identify the colors corresponding to various tissue parts on the part color bar 302 formed by the part position color display control unit, the following are the value settings of various parameters in a preferred embodiment:

[0160] The value range of the ID mark i of the tissue region is [1, 5], the value range of the ID mark k of the tissue part is [0, 24], and the brightness distribution coefficient v is 1 / 10;

[0161] When i=1, that is, the tissue region is an in vitro region, k is 0 or 1, V′ is set to 0.5, H′ and S′ are set to 0, x is set to 0.02, and the saturation distribution value q is set to k, that is,

[0162] V=0.5+p / 10; H=0; S=0.02k;

[0163] When i=2, that is, the tissue region is the esophagus region, k is 2 or 3, V′ is set to 0.75, H′ is set to 15, S′ is set to 0.35, x is set to 0.1, and the saturation distribution value q is set to (k-2), that is,

[0164] V=0.75+p / 10; H=15; S=0.35+0.1*(k-2);

[0165] When i=3, that is, the tissue region is the stomach region, k is set to 4 to 14, V′ is set to 0.65, H′ is set to 5, S′ is set to 0.26, x is set to 0.02, and the saturation distribution value q is set to (k-4), that is,

[0166] V=0.65+p / 10; H=5; S=0.26+0.02*(k-4);

[0167] When i=4, that is, the tissue region is the small intestine region, k is set to 15 to 19, V′ is set to 0.75, H′ is set to 50, S′ is set to 0.35, x is set to 0.03, and the saturation distribution value q is set to (k-15), that is,

[0168] V=0.75+p / 10; H=50; S=0.35+0.03*(k-15);

[0169] When i=5, that is, the tissue region is the large intestine region, k is set to 20 to 24, V′ is set to 0.55, H′ is set to 30, S′ is set to 0.6, x is set to 0.02, and the saturation distribution value q is set to (k-20), that is,

[0170] V=0.55+p / 10; H=30; S=0.6+0.02*(k-20).

[0171] In this way, since the initial hue values ​​H′ set for the detection images in different tissue areas are different, the hue values ​​H calculated from the detection images in different tissue areas are different. At the same time, since the saturation distribution values ​​q set for the detection images in different tissue parts within the same tissue area are different, the saturation values ​​S calculated from the detection images in different tissue parts within the same tissue area are different, which makes it convenient for doctors to distinguish and identify the colors corresponding to the detection images in different tissue parts within the same tissue area on the part color bar 302 formed by the part position color display control unit. Therefore, the part color bar 302 formed by the part position color display control unit is intuitive, highly recognizable, has small errors, and is highly efficient.

[0172] As described above, first, the tissue part and the corresponding tissue area where the shooting position of the detection image is located in the human body are detected by the part position information detection unit, and then the part position color information detection unit calculates the part position color information corresponding to the detection image (including brightness value V, hue value H and saturation value S), and finally the part position color display control unit displays the part position color information corresponding to the time on the scale axis to form a part color bar 302, so that the tissue part where the shooting position of the detection image in the human body is located is displayed in the form of a part color bar 302, so that the doctor can directly know the tissue part where the shooting position of the detection image in the human body is located by identifying the color on the part color bar 302. The part color bar 302 in the present invention is simple and easy to implement, and has high accuracy, fast speed and high efficiency in retrieving and checking detection images, and is easy to identify different tissue parts.

[0173] It should be noted that the same tissue region may correspond to multiple different tissue parts. In order to accurately and quickly locate the detection image corresponding to a certain tissue part on the scale axis, it is preferred to first locate the segmented region corresponding to the tissue part on the scale axis.

[0174] Further, regarding the segmented position information detection unit, the segmented position information detection unit obtains the first tissue part to which the detection image output by the first tissue part detection unit belongs; the segmented position information detection unit calculates the segmented critical points of the first tissue part based on the first tissue part using a segmented position image recognition model; the segmented position information detection unit divides the first tissue part to which the detection image belongs according to the segmented critical points to obtain the segmented area to which the detection image belongs.

[0175] Specifically, the detection image input by the input unit is input into the part position image recognition model for image recognition. Assuming that several first tissue part recognition results are {A, A,…, A, B, B,…, B, C,…, C}, the first tissue part recognition results are then input into the segmented position image recognition model. The segmented position image recognition model analyzes its position distribution through statistical distribution laws, determines that the dividing points of the segmented areas are the positions where A, B and C appear for the last time, and segments the first tissue part to obtain the segmented areas to which the detection image belongs.

[0176] In this way, demarcation and segmentation are performed based on the recognition result of the first tissue part to obtain segmented areas, which are convenient for doctors to view and analyze.

[0177] Furthermore, the segment position color information detection unit sets the segment position color information corresponding to the segment area to which the detection image belongs according to the segment area to which the detection image belongs output by the segment position information detection unit.

[0178] Furthermore, the segmented position color display control unit controls so that the colors corresponding to the segmented position color information corresponding to the segmented area to which the detection image belongs, which is set by the segmented position color information detection unit, are displayed at positions corresponding to the time on the scale axis, respectively, to form a segmented color bar 303.

[0179] Preferably, the segmented position color information corresponding to the segmented area is pre-set, that is, the color and color order on the segmented color bar 303 are fixed. The only difference is that the positions of the dividing points between the colors on the segmented color bar 303 displayed for capsule endoscopy examinations on different subjects may be different.

[0180] Preferably, the pre-set segmented position color information corresponding to the segmented area can be used as the initial brightness value V', initial hue value H', and initial saturation value S' of the first tissue area. For example, the color of the stomach area on the segmented color bar 303 can be pre-set to blue, and the brightness value, hue value, and saturation value of the blue can be used as the initial brightness value V', initial hue value H', and initial saturation value S' of the first tissue area. The calculated color of the part color bar 302 corresponding to the stomach tissue parts such as the fundus and cardia is light blue, dark blue, or other colors close to blue. In this way, the colors corresponding to the part color bar 302 and the segmented color bar 303 at the same position on the scale axis are similar, so that doctors can easily compare and analyze the part color bar 302 and the segmented color bar 303 when reading the film.

[0181] Therefore, when a doctor wants to quickly locate the detection image in tissue part A while reading an image, he can first locate the segmented area B corresponding to the tissue part A according to the different colors on the segmented color bar 303, and then locate the tissue part A according to the different colors on the part color bar 302 within the scale axis range corresponding to the segmented area B. Finally, move the ruler on the scale axis to the scale axis range corresponding to the tissue part A, and the detection image in the tissue part A can be displayed.

[0182] Regarding the first tissue region and the segmented region, it should be noted that in the process of tissue part identification of the detection image, if the results of the part position image recognition model are correct, then the first tissue region to which a certain detection image detected by the part position information detection unit belongs and the segmented region to which the detection image detected by the segmented position information detection unit belongs can be understood as pointing to the same region; however, when the results of the part position image recognition model are incorrect, the first tissue region and the segmented region corresponding to the detection image may point to different regions.

[0183] For example, suppose a certain detection image corresponds to the tissue part of the descending duodenum, but after detection by the part position information detection unit, it is mistakenly identified as the fundus of the stomach. At this time, according to the mapping relationship, it can be known that the first tissue area corresponding to the tissue part (fundus of the stomach) is the stomach, but the segmented position information detection unit performs statistical calculations based on the first tissue part identification result, and can obtain that the tissue part should be located in the segmented area of ​​the small intestine. At this time, the first tissue area and the segmented area point to different areas. The first tissue area is an erroneous detection result, and the segmented area is a correct detection result. The part position detection result of the detection image is finally displayed in the form of a part color bar 302. Since the misidentification of the part position image recognition model is accidental, a small number of part position misidentifications of the part position image recognition model will not have a significant impact on the formation of the part color bar 302, will not affect the doctor's operation and use, and no correction is required.

[0184] The segmented position detection result of the detection image is ultimately displayed in the form of a segmented color bar 303. Since it is necessary to know the accurate dividing point between each segmented area corresponding to the detection image in order to obtain an accurate segmented color bar 303, a segmented position misidentification result of a detection image will have a significant impact on the division of the segmented area (i.e., have a significant impact on the formation of the segmented color bar 303). If the first tissue area corresponding to the first tissue part corresponding to the detection image obtained by the position image recognition model is directly used to perform segmented area boundary division, the misidentification result of the position image recognition model will cause a significant error in the boundary division of the segmented area, thereby obtaining an erroneous segmented color bar 303. Therefore, it is necessary to set a segmented position detection unit to find the dividing point of the accurate segmented area to correct the first tissue area misidentification in the position detection unit, thereby obtaining the correct segmented area boundary division through detection by the segmented position detection unit, and finally obtaining the correct segmented color bar 303.

[0185] In this way, the present invention can correct the misidentification situation that occurs in the part position information detection unit through the segmented position information detection unit, so that the segmented color bar 303 obtained by the segmented position color information detection unit and the segmented position color display control unit is simple and easy to implement, and the retrieval and review of the detection image is accurate, fast, and efficient, and different segmented areas can be easily identified.

[0186] When searching or reviewing detection images directed to different tissue sites, it is also possible to further determine whether the detection images contain lesions.

[0187] Furthermore, the detection unit further includes a lesion information detection unit. The lesion information detection unit inputs the detection images input by the input unit into the image lesion recognition model to perform lesion information detection, wherein the lesion information includes whether a lesion area exists in the detection image.

[0188] Specifically, the lesion information detection unit inputs the detection image input by the input unit into the image lesion recognition model respectively to obtain a second probability value of whether there is a lesion area in the detection image; compares whether the second probability value of the detection image is within a preset probability range; if so, detects that there is a lesion area in the detection image.

[0189] Specifically, the preset probability range is [0.5, 1], that is, when the second probability value of the detection image is between (0, 0.5), the detection image is detected as a normal image and there is no lesion area; when the second probability value is between [0.5, 1], the detection image is detected as an abnormal image and there is a lesion area.

[0190] Furthermore, the control unit also includes a marking point display control unit for controlling to display marking points 304 at positions corresponding to time on the scale axis for the detection image of the lesion area detected by the lesion information detection unit.

[0191] Furthermore, the marking point 304 is formed below the segmented color bar 303 .

[0192] In this way, the lesion information of the detection image can be displayed in the form of a marking point 304 below the scale axis. When the doctor is reading the image, he moves the ruler 301 on the scale axis 305 to the scale axis position corresponding to the marking point 304, and can quickly locate and check the detection image of the lesion area, so as to make a timely and effective diagnosis.

[0193] It can be understood that the image display method and the image display device in the present application can be implemented independently or based on each other, and the corresponding descriptions and configurations can be implemented with each other, which will not be repeated here.

[0194] It should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each implementation method can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0195] The series of detailed descriptions listed above are only specific descriptions of feasible implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any equivalent implementation methods or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. An image display method, characterized in that: The method comprises: An input step of inputting detection images taken by a capsule endoscope in time series; A scale axis display control step, controlling to display the scale axis and displaying a ruler movable on the scale axis, wherein the scale axis represents the entire shooting period of the detection image shot in time series input in the aforementioned input step; An image display control step is performed to control the movement of the ruler on the scale axis and the switching of the image displayed on the image display unit so as to display the detection image at the shooting time corresponding to the position of the ruler on the image display unit; a part position information detection step, inputting the detection images inputted in the input step into a part position image recognition model to detect tissue part information, wherein the tissue part information includes a first tissue part to which the detection image belongs, a first probability value p of the detection image belonging to the first tissue part, and a first tissue region to which the first tissue part belongs; a part position color information detection step, calculating the part position color information of the detection image according to the tissue part information of the detection image detected in the part position information detection step; A part position color display control step, controlling so that the colors corresponding to the part position color information of the detection image calculated by the part position color information detection step are displayed at positions corresponding to the time on the scale axis, respectively, to form a part color bar; A segmented position information detection step, inputting the first tissue part to which the detection image outputted from the part position information detection step belongs into a segmented position image recognition model to perform segmented position information detection, wherein the segmented position information includes the segmented area to which the detection image belongs; A segmentation position color information detection step, according to the segmentation area to which the detection image belongs outputted from the segmentation position information detection step, setting the segmentation position color information corresponding to the segmentation area to which the detection image belongs; The segmented position color display control step controls so that the colors corresponding to the segmented position color information corresponding to the segmented area to which the detection image belongs, which is set by the segmented position color information detection step, are displayed at the positions corresponding to the time on the scale axis, respectively, to form a segmented color bar.

2. The image display method according to claim 1, characterized in that: The location color bar is formed above the scale axis, and the segment color bar is formed below the scale axis.

3. The image display method according to claim 1, characterized in that: The part position information detection step specifically includes: The detection image input in the input step is input into the part position image recognition model, and several probability values ​​of the detection image belonging to any tissue part are calculated; the several probability values ​​are traversed and screened, and the highest probability value is taken as the first probability value p, and the tissue part corresponding to the first probability value p is taken as the first tissue part to which the detection image belongs.

4. The image display method according to claim 1, characterized in that: The part position information detection step specifically includes: The first tissue part to which the detection image belongs is input into a mapping table of human tissue part-tissue area to obtain the first tissue area to which the detection image belongs.

5. The image display method according to claim 1, characterized in that: The part position color information of the detected image calculated according to the part position color information detection step includes a brightness value V, a hue value H and a saturation value S; The step of calculating the brightness value V corresponding to the detected image includes: Determine the initial brightness corresponding to the detection image according to the first tissue region to which the detection image belongs Value V′; According to the brightness distribution coefficient v corresponding to the detection image and the first probability value p, an additional brightness value V" corresponding to the detection image is calculated, and the additional brightness value V" corresponding to the detection image is equal to p*v; The brightness value V corresponding to the detection image is calculated according to the initial brightness value V′ and the additional brightness value V″ corresponding to the detection image, and the brightness value V corresponding to the detection image=V′+V″; The step of calculating the hue value H corresponding to the detected image includes: Determine an initial hue value H′ corresponding to the detection image according to the first tissue region to which the detection image belongs; The hue value H corresponding to the detection image is calculated according to the initial hue value H′ corresponding to the detection image, and the hue value H corresponding to the detection image=H′; The step of calculating the saturation value S corresponding to the detected image includes: Determine an initial saturation value S′ corresponding to the detection image according to the first tissue region to which the detection image belongs; Determine a position coefficient x corresponding to the detection image according to the first tissue region to which the detection image belongs; Determine, according to the first tissue site to which the detection image belongs, a saturation distribution value q corresponding to the detection image; According to the position coefficient x and the saturation distribution value q corresponding to the detection image, the additional saturation value S" corresponding to the detection image is calculated, and the additional saturation value S" corresponding to the detection image is equal to x*q; The saturation value S corresponding to the detection image is calculated based on the initial saturation value S′ and the additional saturation value S″ corresponding to the detection image, and the saturation value S corresponding to the detection image = S′+S″.

6. The image display method according to claim 5, characterized in that: The initial brightness value V′, initial hue value H′, initial saturation value S′, and position coefficient x corresponding to all detection images taken in the same tissue area are the same; the brightness distribution coefficient v corresponding to all detection images is the same; and the saturation distribution value q corresponding to all detection images taken in the same tissue part is the same.

7. The image display method according to claim 1, characterized in that: The segment position information detection step specifically includes: Acquire the first tissue part to which the detection image outputted by the part position information detection step belongs; According to the first tissue part, using a segmented position image recognition model to calculate a segmented critical point of the first tissue part; The first tissue part to which the detection image belongs is segmented according to the segmentation critical point to obtain the segmented area to which the detection image belongs.

8. The image display method according to claim 1, characterized in that: The image display method further includes: a lesion information detection step, inputting the detection images inputted in the input step into the image lesion recognition model to perform lesion information detection, wherein the lesion information includes whether there is a lesion area in the detection image; The marking point display control step controls to display marking points at positions corresponding to time on the scale axis for the detection image of the lesion area detected by the lesion information detection step.

9. The image display method according to claim 8, characterized in that: The lesion information detection step specifically includes: The detection images input by the input step are respectively input into the image lesion recognition model to obtain a second probability value of whether there is a lesion area in the detection image; compare whether the second probability value of the detection image is within a preset probability range; if so, detect that there is a lesion area in the detection image.

10. The image display method according to claim 8, characterized in that: The marking dots are formed below the segmented color bar.

11. An image display device, characterized in that: The image display device comprises: an input unit, a scale axis display control unit, an image display control unit, a part position information detection unit, a part position color information detection unit, a part position color display control unit, a segment position information detection unit, a segment position color information detection unit, and a segment position color display control unit; The input unit inputs detection images taken by the capsule endoscope in time series; The scale axis display control unit controls to display the scale axis and to display a ruler movable on the scale axis, wherein the scale axis represents the entire shooting period of the detection image shot in time series input by the input unit; The image display control unit controls so that the movement of the ruler on the scale axis is synchronously linked with the switching of the detection image displayed in the image display unit, and the detection image at the shooting time corresponding to the position of the ruler is displayed on the image display unit; The part position information detection unit inputs the detection images input by the input unit into the part position image recognition model to perform tissue part information detection, wherein the tissue part information includes a first tissue part to which the detection image belongs, a first probability value p of the detection image belonging to the first tissue part, and a first tissue region corresponding to the first tissue part; The part position color information detection unit calculates the part position color information of the detection image according to the tissue part information of the detection image detected by the part position information detection unit; The part position color display control unit controls so that the colors corresponding to the part position color information of the detection image calculated by the part position color information detection unit are displayed at positions corresponding to time on the scale axis to form a part color bar; The segmented position information detection unit inputs the first tissue part to which the detection image output by the part position information detection unit belongs into the segmented position image recognition model to perform segmented position information detection, wherein the segmented position information includes the segmented area to which the detection image belongs; The segment position color information detection unit sets the segment position color information corresponding to the segment area to which the detection image belongs according to the segment area to which the detection image belongs output by the segment position information detection unit; The segmented position color display control unit controls so that the colors corresponding to the segmented position color information corresponding to the segmented area to which the detection image belongs, which is set by the segmented position color information detection unit, are displayed at the positions corresponding to the time on the scale axis, respectively, to form segmented color bars.

12. The image display device according to claim 11, characterized in that: The location color bar is formed above the scale axis, and the segment color bar is formed below the scale axis.

13. The image display device according to claim 11, characterized in that: The part position information detection unit includes a first tissue part detection unit, which can input the detection image input by the input unit into a part position image recognition model to calculate a number of probability values ​​that the detection image belongs to any tissue part; the first tissue part detection unit traverses and screens the several probability values, takes the highest probability value as the first probability value p, and takes the tissue part corresponding to the first probability value p as the first tissue part to which the detection image belongs.

14. The image display device according to claim 11, characterized in that: The part position information detection unit includes a first tissue region detection unit, which can input the first tissue part to which the detection image belongs detected by the part position information detection unit into a mapping table of human tissue part-tissue region to obtain the first tissue region to which the detection image belongs.

15. The image display device according to claim 11, characterized in that: The part position color information of the detected image includes a brightness value V, a hue value H and a saturation value S, and the part position color information detection unit includes a brightness value detection unit, a hue value detection unit, and a saturation value detection unit; The brightness value detection unit determines an initial brightness value V′ corresponding to the detection image according to the first tissue region to which the detection image belongs detected by the part position information detection unit; the brightness value detection unit According to the brightness distribution coefficient v corresponding to the detection image and the first probability value p of the detection image detected by the part position information detection unit, an additional brightness value V" corresponding to the detection image is calculated, and the additional brightness value V" corresponding to the detection image is = p*v; the brightness value detection unit calculates the brightness value V corresponding to the detection image according to the initial brightness value V′ and the additional brightness value V" corresponding to the detection image, and the brightness value V corresponding to the detection image is = V′+V"; The hue value detection unit determines an initial hue value V′ corresponding to the detection image according to the first tissue region to which the detection image belongs detected by the part position information detection unit; the hue value detection unit calculates the hue value H corresponding to the detection image according to the initial hue value H′ corresponding to the detection image, and the hue value H corresponding to the detection image=H′; The saturation value detection unit determines an initial saturation value S′ corresponding to the detection image according to the first tissue region to which the detection image belongs detected by the part position information detection unit; the saturation value detection unit determines a position coefficient x corresponding to the detection image according to the first tissue region to which the detection image belongs detected by the part position information detection unit; The saturation value detection unit determines the saturation distribution value q corresponding to the detection image based on the first tissue site to which the detection image belongs detected by the site position information detection unit; the saturation value detection unit calculates the additional saturation value S" corresponding to the detection image based on the position coefficient x and the saturation distribution value q corresponding to the detection image, and the additional saturation value S" corresponding to the detection image = x*q; the saturation value detection unit calculates the saturation value S corresponding to the detection image based on the initial saturation value S′ and the additional saturation value S" corresponding to the detection image, and the saturation value S corresponding to the detection image = S′+S".

16. The image display device according to claim 15, characterized in that: The initial brightness value V′, initial hue value H′, initial saturation value S′, and position coefficient x corresponding to all detection images taken in the same tissue area are the same; the brightness distribution coefficient v corresponding to all detection images is the same; and the saturation distribution value q corresponding to all detection images taken in the same tissue part is the same.

17. The image display device according to claim 11, characterized in that: The segmented position information detection unit obtains the first tissue part to which the detection image belongs, which is output by the part position information detection unit; the segmented position information detection unit calculates the segmented critical points of the first tissue part based on the first tissue part using a segmented position image recognition model; the segmented position information detection unit divides the first tissue part to which the detection image belongs according to the segmented critical points to obtain the segmented area to which the detection image belongs.

18. The image display device according to claim 11, characterized in that: The image display device also includes a lesion information detection unit and a marking point display control unit; The lesion information detection unit inputs the detection images input by the input unit into the image lesion recognition model to perform lesion information detection, wherein the lesion information includes whether there is a lesion area in the detection image; The marking point display control unit controls to display marking points at positions corresponding to time on the scale axis for the detection image of the lesion area detected by the lesion information detection unit.

19. The image display device according to claim 18, characterized in that: The lesion information detection unit inputs the detection images input by the input unit into the image lesion recognition model respectively to obtain a second probability value of whether a lesion area exists in the detection image; and compares whether the second probability value of the detection image is within a preset probability range; If so, it is detected that there is a lesion area in the detection image.

20. The image display device according to claim 18, characterized in that The marking dots are formed below the segmented color bar.

21. A capsule endoscope system, comprising a capsule endoscope and a data receiving terminal, characterized in that: The capsule endoscope system further comprises the image display device according to claim 11.

Images