Bleeding detection method, bleeding detection system and image processing device

Abstract

The present application relates to a bleeding detection method, a bleeding detection system, and an image processing device, wherein the bleeding detection method includes: obtaining a first image of a target tissue to be detected captured by an endoscope; determining the amount of bleeding in the target tissue based on the image gradient of the first image to be detected; when the amount of bleeding in the target tissue is the target amount, obtaining several frames of second images of the target tissue to be detected captured by the endoscope before and / or after the first image to be detected; determining the severity of bleeding in the target tissue based on information on changes in color contrast between at least two frames of the first target image; the first target image is an image between the first image to be detected and the several frames of the second image to be detected. Through the present application, the problems of low real-time and low accuracy in bleeding detection are solved, and a timely and efficient bleeding point detection method is realized.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a bleeding detection method, a bleeding detection system and an image processing device. Background Art

[0002] The endoscope is a commonly used medical device. After the insertion part of the endoscope enters the target tissue of the object to be inspected, the endoscope can be used to collect the original image of the target tissue, and the object to be inspected can be inspected or treated based on the original image of the target tissue. However, during the operation, it is very likely that the deep blood vessels under the mucosa will be damaged, resulting in severe bleeding in the target tissue. At this time, if the bleeding cannot be detected and stopped in time, it may cause certain harm to the user. In related technologies, although intraoperative bleeding will be detected, there is a general problem that it is difficult to detect the presence of bleeding in a timely manner, resulting in low real-time and accuracy of bleeding detection.

[0003] Currently, no effective solution has been proposed to address the problems of low real-time and accuracy in bleeding detection in related technologies. Summary of the Invention

[0004] The embodiments of the present application provide a bleeding detection method, a bleeding detection system, and an image processing device to at least solve the problems of low real-time performance and accuracy of bleeding detection in related technologies.

[0005] In a first aspect, an embodiment of the present application provides a method for detecting bleeding, the method comprising:

[0006] Acquiring a first image of the target tissue to be detected collected by the endoscope;

[0007] determining the amount of bleeding of the target tissue according to the image gradient of the first image to be detected;

[0008] When the bleeding amount of the target tissue is the target bleeding amount, acquiring several frames of second images to be detected of the target tissue captured by the endoscope before and / or after the first image to be detected;

[0009] The severity of bleeding of the target tissue is determined based on information on a change in color contrast between at least two frames of first target images; the first target image is an image among the first image to be detected and the plurality of frames of second images to be detected.

[0010] In some embodiments, the method further comprises:

[0011] When the image gradient is less than a preset gradient change threshold, the bleeding condition of the target tissue is determined to be the target bleeding amount condition.

[0012] In some embodiments, determining the severity of bleeding of the target tissue based on information about a change in color contrast between at least two frames of the first target image includes:

[0013] Performing color space conversion processing on each frame of the first target image to obtain a LAB image in a LAB color space and characteristic color dimension information of the LAB image;

[0014] The color contrast change information between each frame of the LAB images is determined according to the characteristic color dimension information, and the bleeding severity is determined according to the color contrast change information between the LAB images.

[0015] In some embodiments, determining the severity of bleeding of the target tissue based on information about a change in color contrast between at least two frames of the first target image includes:

[0016] generating first color change trend information corresponding to the second target image according to a color contrast between the first image to be detected and the second target image in the plurality of frames of second images to be detected;

[0017] determining bleeding volume information of the target tissue according to the first color change trend information;

[0018] The severity of bleeding of the target tissue is determined according to the bleeding volume information and the color contrast change information between the at least two frames of the first target image.

[0019] In some embodiments, determining the bleeding volume information of the target tissue based on the first color change trend information includes:

[0020] determining a first color contrast to be measured corresponding to the number of target pixels in the second target image according to the first color change trend information;

[0021] When it is detected that the first color contrast to be measured is within a preset color area, first concentration information between all the first color contrasts to be measured is calculated, and the bleeding volume information of the target tissue is determined according to the first concentration information.

[0022] In some embodiments, the second target image is the latest frame image among the first image to be detected and the plurality of frames of second images to be detected.

[0023] In some embodiments, determining the severity of bleeding of the target tissue based on information about a change in color contrast between at least two frames of the first target image includes:

[0024] generating second color change trend information corresponding to the first target image according to the color contrast of each frame of the first target image;

[0025] determining the color contrast change information based on change information between all the second color change trend information, and determining the bleeding speed information of the target tissue based on the color contrast change information;

[0026] The bleeding severity of the target tissue is determined according to the bleeding rate information.

[0027] In some embodiments, determining the color contrast change information based on change information between all the second color change trend information includes:

[0028] determining, according to the second color change trend information, a second color contrast to be measured corresponding to the number of target pixels in each frame of the first target image;

[0029] When it is detected that the second color contrast to be measured is within a preset color area, calculating second concentration degree information between all the second color contrasts to be measured in each frame of the first target image;

[0030] The color contrast change information is determined according to the change information between the second concentration degree information corresponding to each frame of the first target image.

[0031] In some embodiments, the image acquisition mode of the endoscope includes a white light mode and a non-white light mode, and after determining the bleeding severity of the target tissue, the method further includes:

[0032] When the bleeding severity is the target severity, the image acquisition mode is switched from the white light mode to the non-white light mode, and a non-white light image of the endoscope in the non-white light mode is acquired; wherein the light irradiated by the endoscope to the target tissue in the white light mode is white light, and the light irradiated to the target tissue in the non-white light mode is non-white light;

[0033] A bleeding position detection result for the target tissue is generated according to the non-white light image.

[0034] In some embodiments, after determining the severity of bleeding in the target tissue, the method further comprises:

[0035] generating bleeding position marking information for the target tissue based on the bleeding position detection result;

[0036] An image to be displayed is generated according to the bleeding position mark information and the white light image.

[0037] In a second aspect, an embodiment of the present application provides a bleeding detection system, the system comprising: a first acquiring unit, a first determining unit, a second acquiring unit, and a second determining unit;

[0038] The first acquiring unit is configured to acquire a first image of the target tissue to be detected acquired by the endoscope;

[0039] The first determining unit is configured to determine the amount of bleeding in the target tissue according to the image gradient of the first image to be detected;

[0040] The second acquiring unit is configured to acquire, when the bleeding amount of the target tissue is a target bleeding amount, a plurality of frames of second images to be detected of the target tissue acquired by the endoscope before and / or after the first image to be detected;

[0041] The second determining unit is used to determine the severity of bleeding of the target tissue based on the change information of the color contrast between at least two frames of first target images; the first target image is an image in the first image to be detected and the plurality of frames of second images to be detected.

[0042] In a third aspect, an embodiment of the present application provides an image processing device, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the bleeding detection method described in the first aspect above.

[0043] In a fourth aspect, an embodiment of the present application provides an endoscope system, which includes: a display, an endoscope, a light source device, and an image processing device as described in the third aspect above.

[0044] Compared with the related art, the bleeding detection method, bleeding detection system and image processing device provided in the embodiments of the present application obtain a first image to be detected of the target tissue collected by an endoscope; determine the bleeding amount of the target tissue based on the image gradient of the first image to be detected; when the bleeding amount of the target tissue is the target bleeding amount, obtain several frames of second images to be detected of the target tissue collected by the endoscope before and / or after the first image to be detected; determine the severity of bleeding of the target tissue based on the change information of the color contrast between at least two frames of the first target image; the first target image is an image in the first image to be detected and the several frames of the second image to be detected, so that bleeding points can be detected in time when they appear, and the bleeding situation of the bleeding points can be detected in real time, avoiding the phenomenon that the bleeding points cannot be detected in time or the bleeding detection accuracy is low due to untimely processing of the images collected by the endoscope, solving the problems of real-time and low accuracy of bleeding situation detection, and realizing a timely and efficient bleeding point detection method.

[0045] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0047] Figure 1 is a flow chart of a bleeding detection method according to an embodiment of the present application;

[0048] Figure 2 is a schematic diagram of pixel statistics of a first target image or a second target image according to an embodiment of the present application;

[0049] Figure 3 is a flow chart of a bleeding detection method according to a preferred embodiment of the present application;

[0050] Figure 4 is a structural block diagram of a bleeding detection system according to an embodiment of the present application;

[0051] Figure 5 is a structural schematic diagram of an endoscope system according to an embodiment of the present application;

[0052] Figure 6 This is a structural diagram of the interior of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided 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 the present application. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the contents disclosed in the present application, some changes such as design, manufacturing or production based on the technical contents disclosed in the present application are only conventional technical means and should not be understood as the contents disclosed in the present application being insufficient.

[0054] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0055] Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by a person of ordinary skill in the technical field to which this application belongs. The words "one", "a", "the" and the like used in this application do not indicate a limit on quantity and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof used in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The word "multiple" used in this application means greater than or equal to two. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The terms "first", "second", "third" and the like involved in this application are merely used to distinguish similar objects and do not represent a specific ordering of the objects.

[0056] An endoscope is an optical device that enters the human body through a natural orifice or a small incision made during surgery to observe relevant areas. Endoscopes of different lengths and diameters can be used to examine different areas of the human body. Depending on the area of ​​examination, endoscopes can be divided into several types, such as laparoscopes, neuroendoscopes, arthroscopy, and esophagoscopes. During surgery, the submucosal blood vessels of the target tissue to be examined may be damaged due to the operation, resulting in heavy bleeding. The blood that flows out covers more and more of the endoscope's image acquisition area, thus affecting the operator's ability to observe the target tissue during surgery from the image.

[0057] In an optional embodiment, an operator can observe the endoscopic image to determine the bleeding situation and the location of the bleeding point. However, the submucosal layer of the target tissue is rich in blood vessels and difficult for the operator to detect. Especially when the bleeding point bleeds rapidly, the blood may not yet cover a large area of ​​the endoscopic image at the current moment. In this case, it will not be determined that hemostasis is required for the bleeding point of the target tissue. The bleeding point is often not detected until the blood begins to fill a large area of ​​the endoscopic image. This makes it difficult to detect the bleeding situation of the target tissue in a timely and accurate manner, which in turn affects the surgical operation.

[0058] In order to improve the real-time and low accuracy issues of bleeding detection, the bleeding detection method provided in the present application uses artificial intelligence computer vision technology to perform image recognition on the endoscopic image obtained by inserting the endoscope into the target tissue, so as to determine the severity of bleeding in the target tissue based on the image color information corresponding to the endoscopic image. Among them, this method is applicable to the application environment of various medical endoscope systems, for example, a medical endoscope system for gastroscopy or an endoscope system for colonoscopy, etc. Moreover, the method can be applied to the host of the endoscope system, that is, the image processing device. Based on this, the embodiment of the present application does not limit the application environment and execution subject of the method. For example, please refer to Figure 5 The application environment includes an endoscope system comprising a display 52, an endoscope 54, a light source device 56, and an image processing device 58. Based on this, the image processing device 58 processes the first image to be detected and the second image to be detected captured by the endoscope, thereby enabling timely and accurate detection of the bleeding condition and severity of the target tissue.

[0059] This embodiment provides a bleeding detection method. Figure 1 is a flow chart of a bleeding detection method according to an embodiment of the present application. Figure 1 As shown, the process includes the following steps:

[0060] Step S110 : obtaining a first image to be detected of the target tissue captured by the endoscope.

[0061] When using an endoscope to examine the digestive tract of a subject, the insertion portion of the endoscope can be inserted into the digestive tract to be examined. In the embodiments of the present application, the part of the digestive tract to be examined by the endoscope is referred to as the target tissue. For example, when using a colonoscopy to examine the intestine, the insertion portion of the colonoscope is inserted into the intestine of the subject to be examined. In this case, the target tissue includes the intestine. For another example, when using a gastroscope to examine the stomach, the insertion portion of the gastroscope is inserted into the stomach. In this case, the target tissue includes the stomach.

[0062] The first image to be detected is a single-frame endoscopic image of the target tissue captured at a specific moment during the endoscope's examination of the target tissue. The first image to be detected is obtained by capturing information about the interior of the target tissue through the endoscope's head end after the endoscope's insertion portion has penetrated the target tissue.

[0063] Step S120: determining the amount of bleeding of the target tissue according to the image gradient of the first image to be detected.

[0064] The bleeding volume condition refers to whether the target tissue is bleeding or whether the bleeding is severe. For example, the target tissue's potential bleeding volume conditions can be pre-classified into primary, secondary, and tertiary bleeding volume conditions. The primary bleeding volume condition indicates moderate bleeding in the target tissue; the secondary bleeding volume condition indicates mild bleeding in the target tissue; and the tertiary bleeding volume condition indicates minimal or no bleeding in the target tissue. The thresholds for classifying the three bleeding volume conditions can be determined based on the target tissue's actual conditions in practice and are not limited here. It is understood that the pre-classified bleeding volume conditions can be further refined to indicate conditions such as severe bleeding in the target tissue, or the ranges within which each bleeding volume condition falls can be further refined, or a two-category classification, such as bleeding and no bleeding, can be implemented. This is not further elaborated here. Then, based on the image gradient of the first image to be detected, the preset bleeding volume condition range within which the actual bleeding volume condition of the target tissue at a given moment falls is determined.

[0065] Specifically, by counting the grayscale value of each pixel in the first image to be detected and the rate of change of the grayscale value of the entire image in the frame of the first image to be detected, the image gradient of the first image to be detected can be obtained. Next, based on the rate of change of the image grayscale value of the first image to be detected, it can be detected whether the target tissue has bleeding such as blood vessel rupture at the corresponding moment when the first image to be detected is collected; for example, when the image gradient difference within the first image to be detected is detected to be small, it indicates that there is a lot of blood in the frame, causing the grayscale values ​​of each pixel to be consistent, or when the grayscale gradient value of the first image to be detected in the corresponding color channel, such as the R channel, is small, it indicates that the corresponding moment when the first image to be detected is collected may have caused blood vessel rupture, that is, blood traces appear in the first image to be detected at this time, and the amount of blood traces in the picture is positively correlated with the image gradient information, thereby determining the corresponding amount of bleeding.

[0066] It is understood that in this embodiment, the image gradient of at least one frame of the first image to be detected can be detected to determine the amount of bleeding in the corresponding frame. If the image gradient of the first image to be detected indicates a normal condition, for example, the amount of bleeding detected corresponding to the first image to be detected is in the aforementioned level 1 bleeding condition, each frame of image can be continuously acquired and real-time monitoring can be performed through the aforementioned steps. If the image gradient of the first image to be detected indicates an abnormal condition, for example, the amount of bleeding detected corresponding to the first image to be detected is in the aforementioned level 2 bleeding condition or level 3 bleeding condition, further detection of the bleeding condition of the target tissue in subsequent steps is performed based on the first detected image of the frame.

[0067] Step S130 , when the bleeding amount of the target tissue is the target bleeding amount, obtaining several frames of second images to be detected of the target tissue collected by the endoscope before and / or after the first image to be detected.

[0068] The target bleeding volume condition refers to a situation at a specific moment in time where a vascular rupture occurs in the target tissue and the bleeding volume is significant, requiring subsequent treatment. For example, the target tissue bleeding volume condition has been pre-classified into the aforementioned first-level, second-level, and third-level bleeding volume conditions. Since the first-level and second-level bleeding volume conditions indicate the presence of a bleeding point in the target tissue and moderate or mild bleeding, the operator may need to pay attention to the real-time status of the bleeding point in the target tissue during surgery. Therefore, the first-level and second-level bleeding volume conditions can be preset as the target bleeding volume conditions. If, based on the first image to be detected, the target tissue bleeding volume condition is detected as being in the first-level or second-level bleeding volume condition through the aforementioned steps, the corresponding bleeding volume condition is determined to be the target bleeding volume condition; accordingly, the third-level bleeding volume condition is preset as a non-target bleeding volume condition. Alternatively, depending on actual application, bleeding volume conditions of other levels, such as a bleeding volume condition indicating severe bleeding in the target tissue, can be preset as the target bleeding volume condition. Alternatively, the above-mentioned three levels of bleeding volume conditions may be further divided, and the bleeding volume conditions in the three levels of bleeding volume conditions indicating that the target tissue is in the range of 5% to 10% of the total blood volume of the body may also be preset as the above-mentioned target bleeding volume conditions; correspondingly, the conditions in the three levels of bleeding volume conditions indicating that the target tissue is not bleeding may be preset as non-target bleeding volume conditions.

[0069] When the target bleeding volume of the target tissue is detected, it indicates that the bleeding volume at the corresponding moment is large, and further, more accurate detection of its specific situation is required. Specifically, in practical applications, bleeding is a phenomenon that lasts for a period of time. Therefore, the specific bleeding situation, such as bleeding rate and bleeding volume, can be analyzed based on the blood change information in multiple frames of images. Based on this, when the embodiment of the present application determines that the target bleeding volume exists in the first image to be detected, images captured near the first image to be detected can be obtained to perform information analysis in the time dimension, thereby determining the required bleeding situation data. More specifically, multiple frames of endoscopic images can be acquired in the time period before the time corresponding to the acquisition of the first image to be detected, that is, a second image to be detected before the first image to be detected can be obtained; alternatively, multiple frames of endoscopic images can be acquired in a period of time after the time corresponding to the acquisition of the first image to be detected as the above-mentioned second image to be detected; alternatively, each frame of endoscopic image within a certain time period including the time when the first image to be detected was acquired can be used as the above-mentioned second image to be detected. It can be seen that there is a temporal correlation between the above-mentioned first image to be detected and each frame of the second image to be detected, which facilitates further subsequent precise detection process.

[0070] Step S140 , determining the severity of bleeding of the target tissue based on information on a change in color contrast between at least two frames of first target images; the first target image is an image among the first image to be detected and the plurality of frames of second images to be detected.

[0071] Specifically, multiple frames of images for further detecting the severity of bleeding in the target tissue can be determined from the first image to be detected and the second image to be detected, and used as the first target image. For example, the first image to be detected and at least one of the plurality of second images to be detected can be used as the first target image, or at least two frames of images can be determined from the plurality of second images to be detected as the first target image. It is understood that, among the at least two first target images, the first target images can be adjacent to each other, or the first target images can be separated by a predetermined interval. For example, the first target image and the second target image can be separated by two frames. Or, among the at least two first target images, some of the first target images can be adjacent to each other, while some of the first target images can be separated by a predetermined interval.

[0072] After determining the above-mentioned multiple frames of first target images, the color contrast information in each frame of the first target image is counted, and then the change rate or change degree value of the color contrast between the multiple frames of first target images is calculated, and the numerical information is determined as the change information of the color contrast between the above-mentioned frames of first target images.

[0073] It should be noted that if a bleeding point has appeared in the target tissue during surgery, and the operator has not promptly performed procedures such as hemostasis, the target tissue may continue to bleed for a period of time, resulting in changes in the color contrast between the multiple frames of the first target image during that period. When the rate of change in the color contrast between the multiple frames is rapid or the degree of change is high, such as when the rate of change in the color contrast between two frames reaches 50% or more, and / or the degree of change reaches 50% or more, this indicates that the blood from the bleeding point in the target tissue during the corresponding time period is spreading rapidly, resulting in a large change in the color contrast between the two frames, i.e., the target tissue has a high bleeding rate during that time period. Conversely, when the rate of change in the color contrast between the multiple frames is slow or the degree of change is low, this indicates that the blood from the bleeding point in the target tissue during the corresponding time period is spreading slowly, i.e., the target tissue has a slow bleeding rate during that time period. Therefore, the bleeding rate between the multiple frames can be determined based on the speed of the color contrast change between the multiple frames. Specifically, when a faster bleeding rate is detected, it means that the wound area of ​​the bleeding point of the target tissue is larger and the bleeding of the target tissue is more serious; when a slower bleeding rate is detected, the corresponding bleeding severity of the target tissue is lower, thereby finally determining the bleeding severity of the above-mentioned target tissue.

[0074] Through steps S110 to S140, the image gradient of the first image to be detected is used to preliminarily determine whether the target tissue is bleeding. This allows for timely detection of bleeding at the point of bleeding. If the target bleeding volume is determined, the bleeding velocity of the bleeding point is then detected in real time based on the color contrast changes between the first image to be detected and the first target image in several frames of the second image to be detected. This avoids situations where the bleeding volume is small but the blood spreads rapidly, resulting in blood filling a large area of ​​the endoscopic image and affecting surgical procedures. Thus, real-time bleeding velocity detection can improve the timeliness of bleeding detection. Furthermore, the data processing of the bleeding detection method is relatively simple, requiring a small number of image frames, resulting in a low processing load and high speed, effectively improving the real-time nature of bleeding detection. Furthermore, by performing precise quantitative analysis of the image information captured by the endoscope, namely determining the bleeding volume based on the image gradient and quantitatively detecting the bleeding velocity based on the color contrast changes between each frame, the inaccuracy of bleeding detection, which is often caused by manually observing the image and determining whether hemostasis is necessary based on experience, is avoided, thereby facilitating accurate detection of bleeding in the target tissue. It can be seen that this embodiment solves the problems of low real-time and low accuracy of bleeding detection through the above steps, and realizes a timely and efficient bleeding point detection method.

[0075] In some embodiments, the bleeding detection method further includes the step of determining that the bleeding condition of the target tissue is a target bleeding volume condition when the image gradient is less than a preset gradient change threshold. The image gradient average of the first image to be detected is calculated, and when the gradient is detected to have dropped below the preset gradient change threshold, it is determined that a blood vessel rupture in the target tissue may have occurred during surgery at the current moment, resulting in the presence of blood in the first image to be detected. Thus, the bleeding condition of the target tissue is determined to be a target bleeding volume condition.

[0076] It should be noted that the target bleeding volume can also be adjusted based on the preset gradient change threshold in accordance with actual application circumstances. For example, in actual applications, if the initial detection step detects the initial appearance of a bleeding point in the target tissue, i.e., the amount of bleeding is relatively small, then the subsequent fine detection step can be performed. In this case, the gradient change threshold can be set to the first threshold. It should be further noted that in related art procedures involving endoscopic access to the target tissue, the submucosal layer of the target tissue is rich in blood vessels and difficult to detect, making it difficult to detect the current bleeding point in a timely manner. In this embodiment, real-time monitoring of the gradient of the entire first image to be detected, combined with the first threshold, facilitates timely detection of the presence of a bleeding point in the target tissue at the current moment, allowing for timely implementation of appropriate measures. Alternatively, if the amount of bleeding in the target tissue in the first image to be detected reaches a certain level in the aforementioned step, then the subsequent fine detection step can be performed. In this embodiment, the gradient change threshold can be set to a second threshold, with the second threshold being greater than the first threshold. It is understood that different levels of gradient change thresholds can be set according to actual circumstances, and this will not be further described here.

[0077] Through the above embodiment, the bleeding condition of the target tissue is determined by a preset gradient change threshold, which is conducive to improving the accuracy of bleeding detection of the target tissue. At the same time, by setting different gradient change thresholds, different bleeding amount ranges can be determined, thereby adjusting the sensitivity of the preliminary detection results of the target tissue during the bleeding detection method, making the bleeding detection method more flexible.

[0078] In some embodiments, the method of determining the severity of bleeding of the target tissue based on information about a change in color contrast between at least two frames of the first target image comprises the following steps:

[0079] First, a color space conversion process is performed on each frame of the first target image to obtain a LAB image in the LAB color space and the characteristic color dimension information of the LAB image. It should be noted that through the above steps, the first image to be detected or the second image to be detected collected by the endoscope is usually an original image in RGB format. Since the pixel value range in the R color channel in the RGB color space is relatively small, it is not conducive to ensuring the accuracy of image-based bleeding detection; therefore, in this embodiment, during the precise detection of bleeding conditions of the target tissue, the first target image originally in the RGB color space can be subjected to a color gamut conversion process to convert the first target image from the RGB color space to the LAB color space with a relatively wider red channel range, so that the detection result can be more obvious. Next, for the LAB image in the LAB color space, the color contrast in the red channel is detected, for example, the a value of all pixels in each frame of the LAB image is detected. The a value is used to represent the degree of color from red to yellow, thereby determining the characteristic color dimension information corresponding to each frame of the LAB image.

[0080] Then, based on the characteristic color dimension information, information about the color contrast change between each LAB image is determined, and the bleeding severity is determined based on this information about the color contrast change between the LAB images. Because the degree of color change in the red channel of an image can be more clearly determined in the LAB color space, this helps improve the accuracy of the bleeding severity determination based on the color contrast information between the LAB images. Thus, through the above embodiment, the first target image is subjected to color gamut conversion to obtain a LAB image, and the bleeding severity is determined based on the LAB image, effectively improving the accuracy of bleeding detection.

[0081] In some embodiments, determining the severity of bleeding of the target tissue based on information about a change in color contrast between at least two frames of the first target image includes:

[0082] Step S141 : generating first color change trend information corresponding to the second target image according to the color contrast between the first image to be detected and the second target image in the plurality of frames of second images to be detected.

[0083] The second target image may be one of the frames of the first target image, or the second target image may be a frame different from the first target image and determined from the first image to be detected and the plurality of frames of the second image to be detected. In another embodiment, the second target image is the latest frame of the first image to be detected and the plurality of frames of the second image to be detected, so as to improve the accuracy of the precise detection of target tissue bleeding based on the second target image.

[0084] Next, color contrast trend information is statistically determined based on the correlation between the color contrast of each pixel in the entire second target image and each pixel, thereby obtaining the first color change trend information. Specifically, the first color change trend information indicates the color contrast trend of all pixels in the entire second target image. This first color change trend information can be determined using histogram statistics, curve fitting, or a fitting formula. For example, a histogram of the color contrast of all pixels in the second target image can be statistically calculated, and based on this histogram, trend information of color contrast changes associated with pixels in a specific color channel can be obtained, thereby obtaining the first color change trend information.

[0085] Step S142: Determine the bleeding volume information of the target tissue according to the first color change trend information.

[0086] Because the first color change trend information carries the color contrast change trend of each pixel in the entire second target image, the concentration of color contrast for each pixel within the red chromaticity region can be determined based on the first color change trend, and the area occupied by blood in the second target image at that moment can be calculated. Specifically, a first mapping relationship between color contrast values ​​and the area occupied by blood in the image can be pre-set based on historical data or prior knowledge to determine the actual area occupied by blood in the image. For example, if the statistically determined color contrast concentration accounts for 60% of all color contrast values, the corresponding actual area occupied by blood in the image can be found to be 10% based on the first mapping relationship.

[0087] Then, the bleeding volume information is determined based on the detected blood area in the image. That is, the larger the blood area in the image, the greater the bleeding volume at the target tissue bleeding point. It should be noted that, compared to the bleeding volume information obtained through qualitative analysis in step S120, in this embodiment, the color contrast of the image is statistically analyzed in the above step to quantitatively determine the blood area, thereby quantitatively determining the corresponding bleeding volume. For example, a second mapping relationship between the blood area in the image and the bleeding volume can be pre-set based on historical data or prior knowledge to determine the actual bleeding volume. For example, if the blood area in the image is determined to be 10%, the corresponding actual bleeding volume value of 5 ml can be found using this second mapping relationship. In another embodiment, a mapping relationship between the color contrast concentration level and the bleeding volume value can be determined based on the first and second mapping relationships, and the actual bleeding volume can be calculated based on this mapping relationship.

[0088] Step S143 : determining the severity of bleeding of the target tissue according to the bleeding volume information and the information on the change in color contrast between the at least two frames of the first target image.

[0089] When the amount of bleeding in the target tissue is greater, that is, the area occupied by the blood in the image is larger, or when the color contrast between multiple frames of the first target image changes faster, that is, the blood area in the image expands faster, it indicates that the severity of the bleeding in the target tissue is higher. Therefore, the bleeding severity can be determined by comprehensive analysis based on the bleeding volume detection results and the color contrast change information between the above-mentioned frames. For example, when the actual bleeding volume value is within the bleeding volume range of 30ml to 50ml, the color contrast change information between the frames indicates that the bleeding rate is within the bleeding rate range of A1 ml / min to A2 ml / min, and the corresponding bleeding severity is the first severity level, which indicates that the target tissue has a large amount of bleeding and a fast bleeding rate, that is, severe bleeding. When the actual bleeding volume value is within the bleeding volume range of 15ml to 30ml, the color contrast change information between the frames indicates that the bleeding rate is within the bleeding rate range of A3 ml / min to A1 ml / min, and the corresponding bleeding severity is the second severity level, which indicates that the target tissue has a large amount of bleeding and a moderate bleeding rate.

[0090] Through the above embodiment, the amount of bleeding is further precisely detected by using the color contrast change information of the first image to be detected and the second target image in the second image to be detected, so that a more accurate bleeding amount detection result can be obtained, effectively improving the accuracy of bleeding detection.

[0091] In an exemplary embodiment, the method of determining the bleeding volume information of the target tissue based on the first color change trend information further includes the following steps:

[0092] Based on the first color change trend information, the first color contrast to be measured corresponding to the target number of pixels in the second target image is determined. The number of pixels in the second target image with the same color contrast is counted based on the first color change trend information. Since a larger number of pixels indicates a larger number of image pixels with this color contrast, the area occupied by the color corresponding to this color contrast in the second target image is larger. Therefore, the number of pixels with the highest numerical values ​​among all the pixel numbers can be used as the target pixel number, and the first color contrast to be measured corresponding to this target pixel number is preferentially detected.

[0093] Next, when it is detected that the first color contrast to be measured is within a preset color region, first concentration information between all first color contrasts to be measured is calculated, and the amount of bleeding in the target tissue is determined based on the first concentration information. The aforementioned color region refers to a range of color values ​​corresponding to hemoglobin determined based on the color gamut of the current image. When it is detected that the first color contrast to be measured is within this color region, it indicates that the image region with the highest color contrast in the current image is filled with blood. The first concentration information between the first color contrasts to be measured can then be further detected to obtain a quantitative result of the bleeding area in the target tissue.

[0094] Specifically, taking the LAB color space as an example, Figure 2 Schematic diagram of pixel point statistics of a first target image or a second target image according to an embodiment of the present application. Figure 2 As shown, a fitting curve representing the first color change trend is statistically generated based on the pixel values ​​of each point in the second target image and the corresponding a values ​​of each pixel value, where a represents the degree of color change from red to yellow. The peaks in the fitting curve represent the number of pixels with the highest values ​​at the same color contrast in the second target image, indicating that the color represented by the color contrast at this peak occupies the largest area in the image. Since hemoglobin is red, when the chromaticity value of the color contrast at the peak is detected to be within the aforementioned color range, i.e., the red chromaticity value range, the first integration level information of the first color contrast to be measured can be determined based on the peak width Δa near the peak. The smaller the Δa value, the higher the integration level of the first color contrast to be measured, i.e., the more pixels with color contrasts approaching the first color contrast to be measured, the larger the area occupied by blood in the second target image, and thus the higher the actual bleeding volume. Based on the aforementioned correlation, the specific actual bleeding volume can be determined based on the statistically obtained Δa value. It can be understood that the peak width Δa value can be calculated from the peak width at the 1 / n peak, and n is a positive number greater than 1.

[0095] In some embodiments, determining the severity of bleeding in the target tissue based on the color contrast transformation information between at least two frames of the first target image further includes the following step: generating second color change trend information corresponding to the first target image based on the color contrast of each frame of the first target image. That is, color contrast trend information is statistically determined based on the correlation between the color contrast of each pixel in the entire image of each frame of the first target image and the color contrast of each pixel, thereby obtaining the second color change trend information corresponding to each frame of the first target image; that is, the second color change trend information is used to indicate the color contrast trend information of all pixels in the entire first target image.

[0096] Then, based on the change information between all the second color change trend information, the color contrast change information is determined, and based on the color contrast change information, the bleeding rate information of the target tissue is determined; and based on the bleeding rate information, the severity of the bleeding in the target tissue is determined. Specifically, the faster the changes between the second color change trend information in each frame of the first target image, the faster the blood spread from the bleeding point in the target tissue during that period, and therefore the faster the bleeding rate of the target tissue.

[0097] In an exemplary embodiment, determining the color contrast change information based on the change information between all the second color change trend information further includes the following steps: determining the second color contrast to be measured corresponding to the number of target pixels in each frame of the first target image based on the second color change trend information; the number of target pixels in the first target image may be the number of pixels with the highest numerical values ​​among all the pixels in the first target image. Upon detecting that the second color contrast to be measured is within a preset color region, calculating second concentration information between all the second color contrasts to be measured in each frame of the first target image; and determining the color contrast change information based on the change information between the second concentration information corresponding to each frame of the first target image.

[0098] Specifically, see Figure 2Based on the pixel values ​​of each point in the first target image and the corresponding a values ​​of each pixel value, a fitting curve is generated to represent the second color change trend. The peaks in the fitting curve represent the number of pixels with the highest values ​​at the same color contrast in the first target image. When the chromaticity value of the color contrast at the peaks falls within the aforementioned color range, and the rate of change in the peak width Δa at the 1 / n peak between each frame of the first target image increases, that is, the rate of change in the second integration level information corresponding to each frame of the first target image increases, this indicates that the bleeding rate in the image is increasing. Furthermore, the aforementioned rate of change information can be determined based on the aforementioned correlation. It should be noted that when the change in Δa between frames reaches a preset width change threshold, the image processing device can determine that a cleansing and hemostasis procedure is necessary, and alert the operator via a display screen display or voice prompt.

[0099] Through the above embodiment, by monitoring the second color change trend information of the first target image, the change rate information between each frame is calculated, so that the severity of bleeding can be determined by specific calculated values, realizing a quantitative detection method for the bleeding speed of the target tissue, which is conducive to improving the accuracy of bleeding detection.

[0100] In some embodiments, the image acquisition mode of the endoscope includes a white light mode and a non-white light mode. After determining the severity of bleeding in the target tissue, the bleeding detection method further includes the following steps:

[0101] Step S151, when the severity of the bleeding is the target severity, switching the image acquisition mode from the white light mode to the non-white light mode, and acquiring a non-white light image of the endoscope in the non-white light mode; wherein, the light irradiated by the endoscope to the target tissue in the white light mode is white light, and the light irradiated to the target tissue in the non-white light mode is non-white light.

[0102] Among them, the severity of bleeding of the target tissue can be divided by setting different preset thresholds such as a bleeding speed threshold or a bleeding volume threshold; when the above embodiment detects that the bleeding volume corresponding to the target image is large and / or the bleeding speed is fast, it means that the current target tissue bleeding is more serious, that is, it is determined to be the above target severity. Under normal circumstances, the above endoscope generally captures images in white light mode to obtain the above first image to be detected and the above second image to be detected. When the bleeding severity of the above target tissue is determined to be the target severity, in order for the operator to stop bleeding in time, the image processing device can control the light source device of the endoscope to switch from white light to non-white light, so as to obtain the non-white light image captured in the non-white light mode at the current moment to detect the location of the bleeding point. It can be understood that in the above non-white light mode, the wavelength band of the irradiation light to which the light source device switches can generally be set to, for example, a near-infrared wavelength range of 600nm to 630nm.

[0103] It should be noted that when the bleeding situation of the target tissue is detected to be at the target severity, the switching method for the image acquisition mode of the above-mentioned endoscope may include a manual switching method and an automatic switching method. For example, when the above-mentioned measurement values ​​are monitored in real time to reach the preset threshold, a text message or voice prompt can be displayed on the display interface to remind the operator to clean the blood in the image and stop the bleeding as soon as possible, and the operator can manually switch the illumination light source device of the endoscope to a non-white light source device to realize the image acquisition mode switching; or, when the current bleeding is detected to be severe, the image processing device can automatically switch the image acquisition mode and control the endoscope to irradiate non-white light to the target tissue to obtain the above-mentioned non-white light image.

[0104] Step S152 : generating a bleeding position detection result for the target tissue according to the non-white light image.

[0105] The above-mentioned bleeding location detection method based on non-white light images can be performed by manual detection or automatic detection using an algorithm. For example, after the endoscope captures the non-white light image, the image processing device can send the captured non-white light image to a display, and the operator can manually detect the non-white light image currently displayed on the display to determine the specific location of the bleeding point in the target tissue at the current moment, thereby obtaining the above-mentioned bleeding location detection result.

[0106] Alternatively, the image processing device can directly utilize the hemoglobin absorption rate characteristics of non-white light in this wavelength band to automatically detect the location of bleeding points through an algorithm in the backend, thereby reducing the number of steps required by the operator and improving surgical efficiency. Specifically, when the blood area and area expansion speed in the image reach a threshold, the operator is prompted by voice or interface to clean the blood stain. Simultaneously, the bleeding location is detected by switching the image observation mode back and forth every other frame. The display image remains in white light mode, that is, the previous frame is an image captured in white light mode, and the next frame is an image captured in non-white light mode, with the white light image displayed on the display. Next, the backend automatically determines the bleeding location through pattern matching. In color space, the orange-yellow area is identified as the blood color in a special mode. The location closest to red in this area of ​​the image is found to determine the specific bleeding location. Alternatively, the bleeding point can be automatically detected through a neural network. In other cases, the non-white light image is input into a neural network trained for bleeding point recognition and the bleeding point location is output. Alternatively, other target detection algorithms can be used to automatically detect the specific location of bleeding points in the target tissue in the backend, which will not be discussed in detail here.

[0107] It should be noted that, when the bleeding point location is detected through the above steps and the operator performs a cleaning and hemostasis operation on the bleeding point of the target tissue, the above-mentioned first target image and / or second target image continue to be acquired and monitored in real time. When it is monitored that the proportion of the blood area occupied by the screen area at the current moment drops below the threshold, the image acquisition mode of the endoscope is switched from the non-white light mode to the white light mode through manual control or automatic control, and the endoscope acquisition work and real-time monitoring are continued.

[0108] Through the above steps S151 to S152, when it is detected that the bleeding severity of the target tissue is the target severity, the image acquisition mode of the endoscope is switched from the white light mode to the non-white light mode, and the bleeding point is found by using the collected non-white light image using the absorption rate characteristics of hemoglobin to light in this band, which is beneficial to improving the accuracy and efficiency of bleeding point position detection, and further effectively improving the accuracy and efficiency of bleeding detection.

[0109] In some embodiments, after determining the severity of bleeding in the target tissue, the bleeding detection method further includes the following steps: generating bleeding position marking information for the target tissue based on the bleeding position detection result; and generating an image to be displayed based on the bleeding position marking information and the white light image.

[0110] Specifically, after obtaining a bleeding location detection result for the target tissue through the above-described method embodiment, corresponding bleeding location marking information can be generated based on the specific bleeding location determined based on the non-white light image. This bleeding location marking information can be information such as a prompt box or textual identification information that indicates the location of the bleeding point in the image. Next, the bleeding location marking information determined in the previous frame is superimposed on the same location in the non-white light image of the next captured white light image to obtain the image to be displayed. Because the time interval between each frame is extremely short, the position difference between the previous and next frames of the same part is not significant. Therefore, the bleeding location found in the previous non-white light image and the next white light image are nearly identical, eliminating the need to worry about the indicated bleeding location error. Finally, the image to be displayed can be directly sent to a display for interface display, or the image to be displayed can be first stored in an image processing device. When an operator enters an interactive display command through the display, the stored image to be displayed is sent to the display in response to the detected display command. For example, taking the use of a prompt box to mark the bleeding position information as an example, in the image to be displayed, based on the phenomenon that the image brightness becomes darker as it approaches the center of the bleeding point, changing the brightness of the prompt box in the image can further help the operator locate the bleeding point as accurately as possible and use electrocoagulation to stop bleeding, making it easier to focus the visual focus.

[0111] Therefore, through the above embodiment, the display screen can continue to display the white light image collected in the white light mode, and at the same time, the bleeding position mark information detected by the background based on the non-white light image is superimposed on the white light image, so as to prompt the operator of the specific location of the bleeding point through the white light image, and the color of the displayed image is consistent with the color inside the organ cavity observed by the human body under normal white light, avoiding affecting the operator's operation and effectively improving the accuracy and efficiency of bleeding detection.

[0112] The following describes the details in conjunction with specific embodiments. Figure 3 is a flow chart of a bleeding detection method according to a preferred embodiment of the present application. Figure 3 As shown in the figure, the calibration process includes the following steps:

[0113] Step S301 : monitoring a first image to be detected and a second image to be detected collected by an endoscope.

[0114] Step S302 , determining whether the image monitoring result obtained in the above step reaches a threshold value, if so, executing the subsequent step S303 , if not, continuing to execute the above step S301 .

[0115] Step S303 : The display performs a picture or sound prompt, and outputs a white light image and a non-white light image in alternate frames.

[0116] In step S304 , the display shows the white light image, and the background automatically searches for bleeding points in the non-white light image.

[0117] In step S305 , the operator cleans and stops bleeding at the target tissue bleeding point based on the monitoring results.

[0118] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0119] This embodiment also provides a bleeding detection system for implementing the above-mentioned embodiments and preferred embodiments. Details already described will not be repeated. As used below, terms such as "unit," "subunit," and the like may refer to a combination of software and / or hardware that implements a predetermined function. While the systems described in the following embodiments are preferably implemented using software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0120] Figure 4 is a structural block diagram of a bleeding detection system according to an embodiment of the present application. Figure 4 As shown, the system includes: a first acquisition unit 42, a first determination unit 44, a second acquisition unit 46, and a second determination unit 48. The first acquisition unit 42 is used to acquire a first image of the target tissue to be detected acquired by the endoscope; the first determination unit 44 is used to determine the amount of bleeding of the target tissue based on the image gradient of the first image to be detected; the second acquisition unit 46 is used to acquire, when the bleeding amount of the target tissue is the target bleeding amount, several frames of second images of the target tissue to be detected acquired by the endoscope before and / or after the first image to be detected; the second determination unit 48 is used to determine the severity of bleeding of the target tissue based on information on changes in color contrast between at least two frames of the first target image; the first target image is an image between the first image to be detected and the several frames of the second images to be detected.

[0121] Through the above embodiment, the first determination unit 44 preliminarily determines whether the target tissue has a bleeding amount through the image gradient of the first image to be detected. The second determination unit 48 detects the severity of bleeding through the change information of the color contrast between the first image to be detected and the first target image in several frames of the second image to be detected when it is determined to be the target bleeding amount. Therefore, the bleeding point can be detected in time when it appears, and the bleeding situation of the bleeding point can be detected in real time, avoiding the phenomenon of failure to detect the bleeding point in time or low accuracy of bleeding detection due to untimely processing of the endoscopic image acquisition, solving the problems of real-time and low accuracy of bleeding situation detection, and realizing a timely and efficient bleeding point detection system.

[0122] In some embodiments, the above-mentioned bleeding detection device also includes a mode switching unit; the mode switching unit is used to switch the image acquisition mode from the white light mode to the non-white light mode when the bleeding severity is the target severity, and obtain a non-white light image of the endoscope in the non-white light mode; wherein the light irradiated by the endoscope to the target tissue in the white light mode is white light, and the light irradiated to the target tissue in the non-white light mode is non-white light; the mode switching unit generates a bleeding position detection result for the target tissue based on the non-white light image.

[0123] In some embodiments, the bleeding detection device further includes a display unit, which is configured to generate bleeding position marking information for the target tissue based on the bleeding position detection result; and which generates an image to be displayed based on the bleeding position marking information and the white light image.

[0124] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0125] This embodiment further provides an image processing device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0126] Optionally, the image processing device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0127] Optionally, in this embodiment, the processor may be configured to execute the following steps through a computer program:

[0128] S1, obtaining a first image of a target tissue to be detected collected by an endoscope.

[0129] S2: Determine the amount of bleeding of the target tissue according to the image gradient of the first image to be detected.

[0130] S3, when the bleeding amount of the target tissue is the target bleeding amount, obtaining several frames of second images to be detected of the target tissue collected by the endoscope before and / or after the first image to be detected.

[0131] S4, determining the severity of bleeding of the target tissue based on information on a change in color contrast between at least two frames of first target images; the first target image is an image among the first image to be detected and the plurality of frames of second images to be detected.

[0132] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be repeated here.

[0133] This embodiment also provides an endoscope system, Figure 5 is a structural diagram of an endoscope system according to an embodiment of the present application, such as Figure 5 As shown, the system includes: a display 52, an endoscope 54, a light source device 56, and an image processing device 58 as described in the above embodiment. In another embodiment, the image processing device 58 can also be deployed in a cloud box connected to the endoscope system via a network.

[0134] This embodiment also provides a computer device, which may be a server device. Figure 6 This is a structural diagram of the internal structure of a computer device according to an embodiment of the present application. Figure 6 As shown. The computer device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store a first image to be detected and a second image to be detected. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a bleeding detection method is implemented.

[0135] Those skilled in the art will understand that Figure 6The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0136] In addition, in conjunction with the bleeding detection method in the above embodiments, the present application can provide a storage medium for implementation. The storage medium stores a computer program; when the computer program is executed by a processor, any one of the bleeding detection methods in the above embodiments is implemented.

[0137] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0138] Those skilled in the art should understand that the various technical features of the above-described embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0139] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for detecting bleeding, characterized in that: The method comprises: Acquiring a first image of the target tissue to be detected collected by the endoscope; determining the amount of bleeding of the target tissue according to the image gradient of the first image to be detected; When the bleeding amount of the target tissue is the target bleeding amount, acquiring several frames of second images to be detected of the target tissue captured by the endoscope before and / or after the first image to be detected; The severity of bleeding of the target tissue is determined based on information on a change in color contrast between at least two frames of first target images; the first target image is an image among the first image to be detected and the plurality of frames of second images to be detected.

2. The bleeding detection method according to claim 1, wherein The method further comprises: When the image gradient is less than a preset gradient change threshold, the bleeding condition of the target tissue is determined to be the target bleeding amount condition.

3. The bleeding detection method according to claim 1, wherein The determining the severity of bleeding of the target tissue according to the change information of the color contrast between at least two frames of the first target image includes: Performing color space conversion processing on each frame of the first target image to obtain a LAB image in a LAB color space and characteristic color dimension information of the LAB image; The color contrast change information between each frame of the LAB images is determined according to the characteristic color dimension information, and the bleeding severity is determined according to the color contrast change information between the LAB images.

4. The bleeding detection method according to any one of claims 1 to 3, characterized in that: The determining the severity of bleeding of the target tissue according to the change information of the color contrast between at least two frames of the first target image includes: generating first color change trend information corresponding to the second target image according to a color contrast between the first image to be detected and the second target image in the plurality of frames of second images to be detected; determining bleeding volume information of the target tissue according to the first color change trend information; The severity of bleeding of the target tissue is determined according to the bleeding volume information and the color contrast change information between the at least two frames of the first target image.

5. The bleeding detection method according to claim 4, characterized in that: The determining the bleeding volume information of the target tissue according to the first color change trend information includes: determining a first color contrast to be measured corresponding to the number of target pixels in the second target image according to the first color change trend information; When it is detected that the first color contrast to be measured is within a preset color area, first concentration information between all the first color contrasts to be measured is calculated, and the bleeding volume information of the target tissue is determined according to the first concentration information.

6. The bleeding detection method according to claim 4, characterized in that: The second target image is the latest frame image among the first image to be detected and the plurality of frames of second images to be detected.

7. The bleeding detection method according to any one of claims 1 to 3, characterized in that: The determining the severity of bleeding of the target tissue according to the change information of the color contrast between at least two frames of the first target image includes: generating second color change trend information corresponding to the first target image according to the color contrast of each frame of the first target image; determining the color contrast change information based on change information between all the second color change trend information, and determining the bleeding speed information of the target tissue based on the color contrast change information; The bleeding severity of the target tissue is determined according to the bleeding rate information.

8. The bleeding detection method according to claim 7, characterized in that: The determining the color contrast change information according to the change information between all the second color change trend information includes: determining, according to the second color change trend information, a second color contrast to be measured corresponding to the number of target pixels in each frame of the first target image; When it is detected that the second color contrast to be measured is within a preset color area, calculating second concentration degree information between all the second color contrasts to be measured in each frame of the first target image; The color contrast change information is determined according to the change information between the second concentration degree information corresponding to each frame of the first target image.

9. The bleeding detection method according to any one of claims 1 to 3, characterized in that: The image acquisition mode of the endoscope includes a white light mode and a non-white light mode. After determining the bleeding severity of the target tissue, the method further includes: When the bleeding severity is the target severity, the image acquisition mode is switched from the white light mode to the non-white light mode, and a non-white light image of the endoscope in the non-white light mode is acquired; wherein the light irradiated by the endoscope to the target tissue in the white light mode is white light, and the light irradiated to the target tissue in the non-white light mode is non-white light; A bleeding position detection result for the target tissue is generated according to the non-white light image.

10. The bleeding detection method according to claim 9, characterized in that: After determining the severity of bleeding in the target tissue, the method further includes: generating bleeding position marking information for the target tissue based on the bleeding position detection result; An image to be displayed is generated according to the bleeding position mark information and the white light image.

11. A bleeding detection system, characterized in that: The system includes: a first acquiring unit, a first determining unit, a second acquiring unit, and a second determining unit; The first acquiring unit is configured to acquire a first image of the target tissue to be detected acquired by the endoscope; The first determining unit is configured to determine the amount of bleeding in the target tissue according to the image gradient of the first image to be detected; The second acquiring unit is configured to acquire, when the bleeding amount of the target tissue is a target bleeding amount, a plurality of frames of second images to be detected of the target tissue acquired by the endoscope before and / or after the first image to be detected; The second determining unit is used to determine the severity of bleeding of the target tissue based on the change information of the color contrast between at least two frames of first target images; the first target image is an image in the first image to be detected and the plurality of frames of second images to be detected.

12. An image processing device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the bleeding detection method according to any one of claims 1 to 10.

13. An endoscope system, characterized in that: The system includes: a display, an endoscope, a light source device, and the image processing device according to claim 12.

Images