Image processing method and apparatus

Abstract

The application relates to an image processing method and device, which comprises the following steps: acquiring a contrast video, the contrast video comprising a plurality of contrast images; determining a target region of interest and a target region of non-interest corresponding to each contrast image, and determining first feature change information of the target region of interest and second feature change information of the target region of non-interest; and determining a key frame image according to the first feature change information and the second feature change information. The image processing method provided by the application can improve the efficiency of determining the key frame image.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to an image processing method and apparatus. Background Technology

[0002] Contrast ultrasound, also known as acoustic contrast imaging, is a technique that uses a contrast agent to enhance backscattered echoes, significantly improving the resolution, sensitivity, and specificity of ultrasound diagnosis. Contrast ultrasound helps physicians locate lesions in real time during ultrasound examinations, improving their ability to interpret and diagnose results. During a contrast ultrasound examination, ultrasound video segments are obtained; to quickly locate lesions, keyframe images within the ultrasound video need to be identified.

[0003] In traditional techniques, ultrasound images are analyzed and keyframes are determined by training a neural network model. However, training the neural network model requires manual annotation of the keyframes in the ultrasound images, which is inefficient. Summary of the Invention

[0004] Therefore, it is necessary to provide an image processing method and apparatus that can improve the determination of keyframe images in order to address the above-mentioned technical problems.

[0005] In a first aspect, one embodiment of this application provides an image processing method, the method comprising:

[0006] Acquire contrast imaging video; the contrast imaging video includes multiple frames of contrast imaging images;

[0007] The target region of interest and the target region of non-interest are determined for each frame of the imaging image, and the first feature change information of the target region of interest and the second feature change information of the target region of non-interest are determined.

[0008] The keyframe image is determined based on the first feature change information and the second feature change information.

[0009] In one embodiment, the keyframe images include a first keyframe image, a second keyframe image, a third keyframe image, and a fourth keyframe image. Determining the keyframe images based on first feature change information and second feature change information includes:

[0010] Based on the first feature change information, determine the first keyframe image, the second keyframe image, and the third keyframe image;

[0011] The fourth keyframe image is determined based on the first feature change information and the second feature change information.

[0012] In one embodiment, determining the first keyframe image, the second keyframe image, and the third keyframe image based on the first feature change information includes:

[0013] The first keyframe image is determined based on the starting point of the first change curve corresponding to the first feature change information;

[0014] The second keyframe image is determined based on the peak point of the first change curve corresponding to the first feature change information;

[0015] Based on the starting point of the first change curve corresponding to the first feature change information, the rising point of the first change curve is determined; and based on the rising point of the first change curve, the third keyframe image is determined.

[0016] In one embodiment, there are multiple regions of interest and multiple corresponding first change curves. Based on the first feature change information, a first keyframe image, a second keyframe image, and a third keyframe image are determined, including:

[0017] The frame image corresponding to the average value of the starting points of all the first change curves is determined as the first keyframe image;

[0018] The frame image corresponding to the average of the peak points of all the first change curves is determined as the second keyframe image;

[0019] The frame image corresponding to the average of the rising points of all the first change curves is determined as the third keyframe image.

[0020] In one embodiment, determining the fourth keyframe image based on the first feature change information and the second feature change information includes:

[0021] Identify the first descending curve segment in the first change curve corresponding to the first feature change information;

[0022] Identify the second descending curve segment in the second change curve corresponding to the second feature change information;

[0023] The fourth keyframe image is determined based on the intersection of the first and second descending curve segments.

[0024] In one embodiment, there are multiple regions of interest, and multiple corresponding first change curves. A fourth keyframe image is determined based on the intersection of the first and second descending curve segments, including:

[0025] The frame image corresponding to the average value of the intersection points between all the first and second descending curve segments is determined as the fourth keyframe image.

[0026] In one embodiment, determining first feature change information of the target region of interest and second feature change information of the target region of non-interest includes:

[0027] For each frame of the imaging image, a first mean of the pixel values ​​of all pixels in the target region of interest is determined, and a second mean of the pixel values ​​of all pixels in the target region of non-interest is determined.

[0028] First feature change information is determined based on the first mean of all frames, and second feature change information is determined based on the second mean of all frames.

[0029] In one embodiment, determining first feature change information based on a first mean of all frames, and determining second feature change information based on a second mean of all frames, includes:

[0030] Based on the first mean of all frames, determine the first angiography intensity value corresponding to the target region of interest in all frames, and based on the second mean of all frames, determine the second angiography intensity value corresponding to the target region of non-interest in all frames.

[0031] First feature change information is determined based on the first contrast intensity value of all frames, and second feature change information is determined based on the second contrast intensity value of all frames.

[0032] In one embodiment, determining the target region of interest and the target region of non-interest corresponding to each frame of the imaging image includes:

[0033] Determine the initial region of interest and initial region of non-interest for each frame of the contrast image;

[0034] Based on motion tracking algorithms, the initial regions of interest and initial regions of non-interest are corrected to obtain the target regions of interest and target regions of non-interest.

[0035] Secondly, one embodiment of this application provides an image processing apparatus, the apparatus comprising:

[0036] The acquisition module is used to acquire contrast-enhanced video; the contrast-enhanced video includes multiple frames of contrast-enhanced images.

[0037] The information determination module is used to determine the target region of interest and the target region of non-interest corresponding to each frame of the imaging image, and to determine the first feature change information of the target region of interest and the second feature change information of the target region of non-interest;

[0038] The image determination module is used to determine the keyframe image based on the first feature change information and the second feature change information.

[0039] The aforementioned image processing method and apparatus acquire contrast-enhanced video, which includes multiple frames of contrast-enhanced images; determine the target region of interest (ROI) and target non-ROI corresponding to each frame of contrast-enhanced image, and determine first feature change information of the ROI and second feature change information of the non-ROI; and determine keyframe images based on the first and second feature change information. In this embodiment, keyframe images are directly determined based on the first feature change information of the ROI in each frame of the contrast-enhanced video and the second feature change information of the non-ROI in each frame. Compared with traditional techniques, this method eliminates the need for pre-training a neural network model in determining keyframe images, improving efficiency and reducing hardware requirements, thus making the image processing method more practical. Attached Figure Description

[0040] Figure 1 This is an application environment diagram of an image processing method in one embodiment;

[0041] Figure 2 This is a flowchart illustrating the steps of an image processing method in one embodiment;

[0042] Figure 3 This is a flowchart illustrating the steps of an image processing method in another embodiment;

[0043] Figure 4 This is a flowchart illustrating the steps of an image processing method in another embodiment;

[0044] Figure 5 This is a flowchart illustrating the steps of an image processing method in another embodiment;

[0045] Figure 6 This is a flowchart illustrating the steps of an image processing method in another embodiment;

[0046] Figure 7 This is a flowchart illustrating the steps of an image processing method in another embodiment;

[0047] Figure 8 This is a schematic diagram of the first variation curve in one embodiment;

[0048] Figure 9 This is a flowchart illustrating the steps of an image processing method in another embodiment;

[0049] Figure 10 This is a flowchart illustrating the steps of an image processing method in another embodiment;

[0050] Figure 11 This is a schematic diagram of the structure of an image processing device in one embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0052] The serial numbers assigned to components in this article, such as "first" and "second", are used only to distinguish the objects being described and have no sequential or technical meaning.

[0053] Before detailing the technical solutions of the embodiments disclosed in this application, the background technology or technological evolution on which the embodiments of this application are based will be introduced first. In the medical field, contrast ultrasound (also known as acoustic contrast imaging) is a technique that uses contrast agents to enhance backscattered echoes, significantly improving the resolution, sensitivity, and specificity of ultrasound diagnosis. Contrast ultrasound technology helps physicians locate lesions in real time during ultrasound contrast examinations, improving the ability to interpret and diagnose results. In contrast ultrasound examinations, ultrasound contrast video segments are obtained. To quickly locate lesions, keyframe images within these video segments need to be determined. Currently, there is limited research on keyframe extraction methods for contrast ultrasound. The main research approach involves image analysis of ultrasound images, such as manually extracting image features, learning these features through neural networks, and then determining the keyframe images. In practical applications, due to the diverse morphologies of lesion regions in ultrasound images, corresponding to different types of lesions, neural networks cannot effectively predict keyframe images. Furthermore, training neural networks requires high hardware specifications, and the samples needed for training require manual annotation, resulting in long training cycles and low efficiency. In response, this application provides an image processing method.

[0054] The technical solution of this application and how the technical solution of this application solves the technical problem are described in detail below with specific embodiments.

[0055] The image processing method provided in this application can be applied to computer devices, including but not limited to industrial computers, laptops, and tablets. The internal results of the computer device are as follows: Figure 1As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements an image processing method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad located on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0056] In one embodiment, such as Figure 2 As shown, an image processing method is provided. This embodiment illustrates the application of this method to a computer device. In this embodiment, the method includes the following steps:

[0057] Step 200: Obtain the contrast-enhanced video; the contrast-enhanced video includes multiple frames of contrast-enhanced images.

[0058] Contrast-enhanced videos are obtained by scanning the object being processed using a scanning device. A contrast-enhanced video consists of multiple frames of contrast images. The scanning device can be an ultrasound machine or a CT (Computed Tomography) camera. Contrast-enhanced videos obtained using an ultrasound machine are called ultrasound contrast-enhanced videos, while those obtained using a CT machine are called X-ray contrast-enhanced videos.

[0059] The contrast-enhanced video can be pre-stored in the post-processing workstation of the imaging equipment or in a server such as a PACS (Picture Archiving and Communication Systems). The computer equipment can retrieve the contrast-enhanced video from the post-processing workstation or from the PACS server. Alternatively, the contrast-enhanced video can be pre-stored in the computer equipment's memory, which the computer equipment can retrieve directly when needed. This embodiment does not limit the specific method for retrieving the contrast-enhanced video, as long as the function can be achieved.

[0060] Step 210: Determine the target region of interest and the target region of non-interest corresponding to each frame of the imaging image, and determine the first feature change information of the target region of interest and the second feature change information of the target region of non-interest.

[0061] The region of interest (ROI) in a contrast-enhanced image refers to the lesion area within the image, while the region of non-lesion (NOT) refers to the non-lesion area within the image. A contrast-enhanced image may include one or multiple lesion areas, thus the identified ROI can be one or multiple.

[0062] After acquiring the contrast-enhanced video, the computer device analyzes each frame of the contrast-enhanced image to determine the target region of interest (ROI) and the target region of non-ROI in each frame. This embodiment does not limit the specific method used to determine the ROI and non-ROI in each frame of the contrast-enhanced image, as long as the function is achieved.

[0063] In an optional embodiment, the target region of interest (ROI) and target non-ROI corresponding to each frame of contrast imaging image can be determined through human-computer interaction. That is, each frame of contrast imaging image is displayed on the screen of a computer device; the user's trigger operation on each frame of contrast imaging image is received, and the target ROI and target non-ROI in each frame of contrast imaging image are determined. In other words, the user marks the target ROI and target non-ROI in each frame of contrast imaging image on the screen using curves of different colors.

[0064] In another alternative embodiment, a pre-trained recognition model is stored in the memory of the computer device. After acquiring the contrast-enhanced video, the computer device inputs the contrast-enhanced video into the recognition model. The recognition model can first divide the contrast-enhanced video into multiple contrast-enhanced images, and then perform recognition processing on each contrast-enhanced image to determine the target region of interest and the target region of non-interest in each contrast-enhanced image.

[0065] After determining the target region of interest (ROI) in each frame of the contrast-enhanced image, the computer device analyzes the ROI in each frame to determine the first feature change information of the ROI. This first feature change information characterizes the feature changes of the ROI in the contrast-enhanced video over time. This embodiment does not limit the specific content of the first feature change information, as long as it can characterize the feature changes of the ROI.

[0066] Similarly, after determining the target non-interest region in each frame of the contrast-enhanced image, the computer device analyzes the target non-interest region in each frame to determine the second feature change information of the target non-interest region. The second feature change information is used to characterize the feature changes of the target non-interest region in the contrast-enhanced video over time. This embodiment does not limit the specific content of the second feature change information, as long as it can characterize the feature changes of the target non-interest region.

[0067] Step 220: Determine the keyframe image based on the first feature change information and the second feature change information.

[0068] After determining the first feature change information of the target region of interest and the second feature change information of the target region of non-interest, the computer equipment determines keyframe images in the contrast-enhanced video based on the first and second feature change information. There can be one or more keyframe images in the contrast-enhanced video. A keyframe image refers to the relevant features (position, shape, and size, etc.) of the target region of interest corresponding to the object to be processed in the contrast-enhanced video that can more accurately and clearly characterize it.

[0069] The image processing method provided in this application involves acquiring a contrast-enhanced video, which includes multiple frames of contrast-enhanced images; determining a target region of interest (ROI) and a target region of non-interest (NOI) for each frame of the contrast-enhanced image; determining first feature change information for the ROI and second feature change information for the NOI; and determining keyframe images based on the first and second feature change information. In this embodiment, keyframe images are determined directly based on the first feature change information of the ROI in each frame of the contrast-enhanced video and the second feature change information of the NOI in each frame. Compared to traditional techniques, this method eliminates the need for pre-training a neural network model, improving efficiency and reducing hardware requirements, thus offering greater practicality. Furthermore, the first and second feature change information characterize the changes in features of the ROI and NOI in the contrast-enhanced video over time, enabling clear acquisition of feature changes in each frame of the contrast-enhanced image, thereby improving the accuracy of the determined keyframe images and enhancing the reliability of the image processing method.

[0070] In one embodiment, four keyframe images are determined based on first feature change information and second feature change information, namely, a first keyframe image, a second keyframe image, a third keyframe image, and a fourth keyframe image. In this case, as... Figure 3 As shown, this relates to an implementation method for determining a keyframe image based on first feature change information and second feature change information. The steps of this implementation method include:

[0071] Step 300: Based on the first feature change information, determine the first keyframe image, the second keyframe image, and the third keyframe image.

[0072] The computer device can determine three keyframe images—a first keyframe image, a second keyframe image, and a third keyframe image—based on the first feature change information of the target region of interest. In other words, the computer device can select three keyframe images from the imaging video based on the first feature change information of the target region of interest.

[0073] Step 310: Determine the fourth keyframe image based on the first feature change information and the second feature change information.

[0074] The computer device can determine the fourth keyframe image based on the first feature change information of the target region of interest and the second feature change information of the target region of non-interest.

[0075] In this embodiment, based on the first feature change information of the target region of interest, a first keyframe image, a second keyframe image, and a third keyframe image can be determined to accurately represent the relevant features of the target region of interest. Considering that there may be interaction between the determined target region of interest and the target region of non-interest, based on the first feature change information of the target region of interest and the second feature change information of the target region of non-interest, a fourth keyframe image can be determined to accurately represent the relevant features when there is interaction between the target region of interest and the target region of non-interest. The first, second, third, and fourth keyframe images determined in this way can more accurately and clearly represent the relevant features of the target region of interest in the object to be processed, making the image processing method more reliable.

[0076] In one embodiment, one method for determining a first keyframe image based on first feature change information includes:

[0077] The first keyframe image is determined based on the starting point of the first change curve corresponding to the first feature change information.

[0078] The first feature change information represents the feature changes of the target region of interest over time. The computer device can fit a first change curve based on the obtained first feature change information, where the horizontal axis of the first change curve represents time, and the vertical axis represents the feature information of the target region of interest. This embodiment does not limit the specific process of fitting the first change curve based on the first feature change information, as long as the function can be achieved.

[0079] After determining the first change curve corresponding to the first feature change information, the computer device determines the starting point in the first change curve, that is, the first point appearing in the first change curve. The first keyframe image is then determined based on the starting point of the first change curve.

[0080] In this embodiment, a first change curve corresponding to the first feature change information is first determined, and then a first keyframe image is determined based on the starting point of the first change curve. The first change curve can more intuitively and clearly represent the feature changes of the target region of interest, thereby enabling a more accurate first keyframe image to be determined based on the starting point of the first change curve.

[0081] In an optional embodiment, if there is one target region of interest in the determined imaging image, then there is one first feature change information for the target region of interest, and a corresponding first change curve. In this case, the computer device directly determines the frame image corresponding to the starting point of the first change curve as the first keyframe image, that is, determines the frame image corresponding to the time value at the starting point of the first change curve as the first keyframe image.

[0082] In one embodiment, if there are multiple regions of interest (ROIs) in the determined imaging image, then there are multiple first feature change information corresponding to the ROIs, and thus multiple first change curves. In this case, an implementation method involves determining a first keyframe image based on the starting point of the first change curve corresponding to the first feature change information. This implementation method includes:

[0083] The frame image corresponding to the average value of the starting points of all the first change curves is determined as the first keyframe image.

[0084] When the computer device obtains multiple first change curves, it acquires the starting point of each first change curve, calculates the average value of the starting points of all first change curves, that is, calculates the average value of the time values ​​among the starting points of all first change curves, and determines the frame image corresponding to the average value as the first keyframe image.

[0085] In an optional embodiment, if the starting point of all the first change curves is the same, the frame image corresponding to the starting point of any one of the first change curves can be selected and determined as the first keyframe image.

[0086] In this embodiment, a method for determining the first keyframe image is provided when there are multiple first change curves. This method is simple in logic and easy to implement, making the image processing method more practical.

[0087] In one embodiment, an implementation method for determining a second keyframe image based on first feature change information is designed, the implementation method comprising:

[0088] The second keyframe image is determined based on the peak point of the first change curve corresponding to the first feature change information.

[0089] After obtaining the first feature change information of the target region of interest, the computer device first determines the first change curve corresponding to the first feature change information. The description of the first change curve can be found in the specific description in the above embodiments, and will not be repeated here.

[0090] After determining the first change curve corresponding to the first feature change information, the computer device identifies the peak point in the first change curve, that is, the point with the largest ordinate value in the first change curve. The second keyframe image is then determined based on the peak point of the first change curve.

[0091] In this embodiment, a first change curve corresponding to the first feature change information is first determined, and then the second keyframe image is determined based on the peak point of the first change curve. The first change curve can more intuitively and clearly represent the feature changes of the target region of interest, thereby enabling a more accurate second keyframe image to be determined based on the peak point of the first change curve.

[0092] In an optional embodiment, if there is one target region of interest in the determined imaging image, then there is one first feature change information for the target region of interest, and a corresponding first change curve. In this case, the computer device directly determines the frame image corresponding to the peak point of the first change curve as the second keyframe image, that is, the frame image corresponding to the time value in the peak point of the first change curve is determined as the second keyframe image.

[0093] In one embodiment, if there are multiple target regions of interest in the determined imaging image, then there are multiple first feature change information corresponding to the target regions of interest, and multiple corresponding first change curves. In this case, an implementation method involves determining a second keyframe image based on the peak points of the first change curves corresponding to the first feature change information. This implementation method includes:

[0094] The frame image corresponding to the average of the peak points of all the first change curves is determined as the second keyframe image.

[0095] When the computer device obtains multiple first change curves, it acquires the peak point of each first change curve, calculates the average value of the peak points of all first change curves, that is, calculates the average value of the time values ​​among the peak points of all first change curves, and determines the frame image corresponding to the average value as the second keyframe image.

[0096] In an optional embodiment, if no frame image corresponding to the average of the time values ​​among the peak points of all the first change curves is found, the frame image corresponding to the time value closest to the average value is determined as the third keyframe image.

[0097] In this embodiment, a method for determining the second keyframe image is provided when there are multiple first change curves. This method is simple in logic and easy to implement, making the image processing method more practical.

[0098] In one embodiment, such as Figure 4 As shown, an implementation method for determining a third keyframe image based on first feature change information is described, and the steps of this implementation method include:

[0099] Step 400: Determine the rising point of the first change curve based on the starting point of the first change curve corresponding to the first feature change information.

[0100] After obtaining the first feature change information, the computer device determines the starting point of the first change curve corresponding to the first feature change information. The description of determining the starting point of the first change curve can be found in the specific description in the above embodiments, and will not be repeated here.

[0101] After determining the starting point of the first change curve, the computer device determines the rising point of the first change curve based on the starting point. The rising point is a point in the rising curve segment of the first change curve.

[0102] In an optional embodiment, the computer device can determine the rising point according to the formula A*(1+a%), where A represents the value corresponding to the ordinate of the starting point in the first change curve, and a is a parameter preset by the user.

[0103] Step 410: Determine the third keyframe image based on the rising point of the first change curve.

[0104] After determining the rising point of the first change curve, the computer device determines the third keyframe image based on that rising point.

[0105] In this embodiment, the rising point of the first change curve is first determined based on its starting point, and then the third keyframe image is determined based on this rising point. This method for determining the third keyframe image is logically simple and easy to implement.

[0106] In an optional embodiment, if there is one target region of interest in the determined imaging image, and there is also one corresponding first change curve, then there is one rising point of the determined first change curve. In this case, the computer device directly determines the frame image corresponding to the rising point of the first change curve as the third keyframe image, that is, determines the frame image corresponding to the time value in the rising point of the first change curve as the third keyframe image.

[0107] In one embodiment, if there are multiple regions of interest in the determined imaging image, there are multiple corresponding first change curves, and multiple determined rise points. In this case, an implementation method for determining a third keyframe image based on the rise points of the first change curves is provided, the implementation method including:

[0108] The frame image corresponding to the average of the rising points of all the first change curves is determined as the third keyframe image.

[0109] The computer device can determine the rising point corresponding to all first change curves based on the starting point of each first change curve, calculate the average time value among the rising points corresponding to all first change curves, and determine the frame image corresponding to the average value as the third keyframe image.

[0110] In this embodiment, a method for determining the third keyframe image is provided when there are multiple first change curves. This method is simple in logic and easy to implement, making the image processing method more practical.

[0111] In one embodiment, such as Figure 5 As shown, this relates to an implementation method for determining a fourth keyframe image based on first feature change information and second feature change information. The steps of this implementation method include:

[0112] Step 500: Determine the first descending curve segment in the first change curve corresponding to the first feature change information.

[0113] After obtaining the first feature change information, the computer device determines the first change curve corresponding to the first feature change information. The description of determining the first change curve can be found in the specific description in the above embodiments, and will not be repeated here.

[0114] After obtaining the first change curve, the computer analyzes it to determine the first descending curve segment. The first descending curve segment is the segment of the first change curve where the relevant feature information (i.e., the value of the ordinate) decreases over time.

[0115] Step 510: Determine the second descending curve segment in the second change curve corresponding to the second feature change information.

[0116] The second feature change information represents the feature changes of the target's non-interest region over time. The computer device can fit a second change curve based on the obtained second feature change information, where the horizontal axis of the second change curve represents time, and the vertical axis represents the feature information of the target's non-interest region. This embodiment does not limit the specific process of fitting the second change curve based on the second feature change information, as long as the function can be achieved.

[0117] After determining the second change curve corresponding to the second feature change information, the computer equipment analyzes the second change curve to identify the second descending curve segment. The second descending curve segment is the curve segment in the second change curve where the relevant feature information (i.e., the value of the ordinate) decreases over time.

[0118] Step 520: Determine the fourth keyframe image based on the intersection point between the first and second descending curve segments.

[0119] The computer device fits the first change curve corresponding to the first feature change information and the second change curve corresponding to the second feature change information into the same coordinate system. After obtaining the first descending curve segment of the first change curve and the second descending curve segment of the second change curve, the computer device can determine the intersection point between the first descending curve segment and the second descending curve segment in the same coordinate system, and determine the fourth keyframe image based on the intersection point.

[0120] In this embodiment, the fourth keyframe image is determined based on the intersection point between the first descending curve segment of the first change curve corresponding to the first feature change information and the second descending curve segment of the second change curve corresponding to the second feature change information. This intersection point characterizes the time when the features of the target region of interest and the non-region of interest change, thus allowing the user to accurately analyze the region of interest of the object to be processed using the fourth keyframe image determined by this intersection point. Furthermore, this method of determining the fourth keyframe image is logically simple and easy to implement, improving the practicality of the image processing method.

[0121] In an optional embodiment, if there is one target region of interest in the determined imaging image, and there is also one corresponding first variation curve, then there is also one intersection point between the first descending curve segment of the determined first variation curve and the second descending curve segment of the determined second variation curve. In this case, the computer device directly determines the frame image corresponding to the time value of the intersection point between the first descending curve segment of the first variation curve and the second descending curve segment of the second variation curve as the fourth keyframe image.

[0122] In one embodiment, if there are multiple regions of interest in the determined imaging image, and multiple corresponding first variation curves, then there are also multiple intersection points between the first descending curve segment of the determined first variation curve and the second descending curve segment of the determined second variation curve. In this case, an implementation method for determining a fourth keyframe image based on the intersection points between the first and second descending curve segments is provided, the implementation method including:

[0123] The frame image corresponding to the average value of the intersection points between all the first and second descending curves is determined as the fourth keyframe image.

[0124] After obtaining the first descending curve segment of multiple first change curves and the second descending curve segment of the second change curve, the computer device determines the intersection point between each first descending curve segment and the second descending curve segment, calculates the average value of the time values ​​corresponding to all intersection points, and determines the frame image corresponding to the average value as the fourth keyframe image.

[0125] In this embodiment, a method for determining the fourth keyframe image is provided when there are multiple first change curves. This method is simple in logic and easy to implement.

[0126] In one embodiment, such as Figure 6 As shown, an implementation method involves determining first feature change information of a target region of interest and second feature change information of a target region of non-interest. The steps of this implementation method include:

[0127] Step 600: For each frame of the imaging image, determine the first mean of the pixel values ​​of all pixels in the target region of interest, and determine the second mean of the pixel values ​​of all pixels in the target region of non-interest.

[0128] After acquiring the contrast-enhanced video, the computer device obtains the pixel value of each pixel in the target region of interest for each frame of the contrast-enhanced image, and calculates the average value of all pixel values ​​to obtain a first mean. If the contrast-enhanced image includes multiple target regions of interest, then the contrast-enhanced image corresponds to multiple first means.

[0129] Similarly, for each frame of the imaging image, the pixel value of each pixel in the non-interest region of the target in the imaging image is obtained, and the average value of all pixel values ​​is calculated to obtain the second mean.

[0130] Step 610: Determine the first feature change information based on the first mean of all frames, and determine the second feature change information based on the second mean of all frames.

[0131] After acquiring the first mean value corresponding to the target region of interest in all frames of the contrast-enhanced images, the computer device determines the first feature change information based on the first mean value. The first feature change information is used to characterize the change in pixel value of the target region of interest in the contrast-enhanced video over time.

[0132] After acquiring the second mean value corresponding to the non-interest regions of interest (NIRO) of the target in all frames of the contrast-enhanced images, the computer device determines the second feature change information based on the second mean value. The second feature change information is used to characterize the change in pixel values ​​of the NIRO in the contrast-enhanced video over time.

[0133] In this embodiment, for each frame of the imaging image, the first feature change information is determined by the first average of the pixel values ​​of all pixels in the target region of interest, and the second feature change information is determined by the second average of the pixel values ​​of all pixels in the target region of non-interest. This clearly defines the first feature change information as the change in pixel values ​​in the target region of interest over time, and the second feature change information as the change in pixel values ​​in the target region of non-interest over time. Furthermore, the method for determining the first and second feature change information is quick and easy to implement, improving the practicality of the image processing method.

[0134] When the first feature change information represents the change in pixel values ​​of the target's region of interest in the imaging video over time, the horizontal axis of the first change curve corresponding to the first feature change information represents time, and the vertical axis represents pixel values. Similarly, when the second feature change information represents the change in pixel values ​​of the target's region of non-interest in the imaging video over time, the horizontal axis of the second change curve corresponding to the second feature change information represents time, and the vertical axis represents pixel values.

[0135] In one embodiment, such as Figure 7 As shown, this involves determining first feature change information based on a first mean of all frames, and determining second feature change information based on a second mean of all frames, including:

[0136] Step 700: Based on the first mean of all frames, determine the first imaging intensity value corresponding to the target region of interest in all frames, and based on the second mean of all frames, determine the second imaging intensity value corresponding to the target region of non-interest in all frames.

[0137] After acquiring the first mean value corresponding to the target region of interest in all frames of contrast images, the computer device determines the first contrast intensity value corresponding to the target region of interest in each frame of contrast images based on the first mean value corresponding to each frame of contrast images. That is, the intensity of the contrast agent in the target region of interest in the frame of contrast images, thereby obtaining the first contrast intensity value corresponding to the target region of interest in all frames.

[0138] After acquiring the second mean value corresponding to the target non-interest region in all frames of contrast images, the computer device determines the second contrast intensity value corresponding to the target non-interest region of the contrast image for that frame based on the second mean value corresponding to each frame of contrast images, that is, the intensity of the contrast agent in the target non-interest region of the contrast image for that frame.

[0139] In an optional embodiment, the computer device can be configured according to the formula. Calculate the contrast intensity value. Here, b and c are preset parameters; optionally, b is greater than or equal to 40, and c = 20. Preferably, b = 40, and n is a first mean or a second mean. Inputting n as the first mean in this formula yields a first contrast intensity value; inputting n as the second mean yields a second contrast intensity value.

[0140] Step 710: Determine the first feature change information based on the first contrast intensity value of all frames, and determine the second feature change information based on the second contrast intensity value of all frames.

[0141] After acquiring the first contrast intensity value corresponding to the target region of interest in all frames of contrast-enhanced images, the computer device determines first feature change information based on the first contrast intensity value. The first feature change information is used to characterize the change in the contrast intensity value of the target region of interest in the contrast-enhanced video over time.

[0142] After acquiring the second contrast intensity value corresponding to the non-interest region of the target in all frames of contrast images, the computer device determines the second feature change information based on the second contrast intensity value. The second feature change information is used to characterize the change in the contrast intensity value of the non-interest region of the target in the contrast video over time.

[0143] In this embodiment, for each frame of contrast-enhanced image, a first contrast intensity value is determined by the first average of the pixel values ​​of all pixels in the target region of interest (ROI). First feature change information is then determined based on this first contrast intensity value. A second contrast intensity value is determined by the second average of the pixel values ​​in the target non-ROI region (ROI). Second feature change information is then determined based on this second contrast intensity value. This clearly defines the first feature change information as the change in contrast intensity value of the ROI over time, and the second feature change information as the change in contrast intensity value of the ROI over time. Furthermore, the method for determining the first and second feature change information is quick and easy to implement, improving the practicality of the image processing method.

[0144] When the first feature change information characterizes the change in the region of interest (ROI) of the target in the contrast-enhanced video over time, and the change in contrast intensity value, the horizontal axis of the first change curve corresponding to the first feature change information represents time, and the vertical axis represents the contrast intensity value. The first change curve is shown below. Figure 8As shown. Similarly, when the second feature change information represents the change in contrast intensity value of the non-interest region of the target in the contrast video over time, the horizontal axis of the second change curve corresponding to the second feature change information represents time, and the vertical axis represents the contrast intensity value. For the target region of interest and the target non-interest region, the contrast agent has the characteristic of rapid entry and exit, and the target region of interest is more sensitive to the contrast agent. Therefore, the contrast intensity value of the first change curve is higher than that of the second change curve. The intersection of the first descending curve segment of the first change curve and the second descending curve segment of the second change curve can more clearly distinguish the first change curve and the second change curve. In this way, the fourth keyframe image determined by the intersection of the first descending curve segment and the second descending curve segment is more accurate, which makes it easier for users to accurately analyze the region of interest of the object to be processed.

[0145] In one embodiment, such as Figure 9 As shown, this relates to an implementation method for determining the target region of interest and the target region of non-interest corresponding to each frame of angiographic image. The steps of this implementation method include:

[0146] Step 900: Determine the initial region of interest and the initial region of non-interest corresponding to each frame of the imaging image.

[0147] After acquiring the contrast-enhanced video, the computer equipment determines the initial region of interest and the initial region of non-interest in each frame of the contrast-enhanced image.

[0148] In an optional embodiment, the computer device can determine the initial region of interest (ROI) and initial non-ROI in each frame of contrast imaging image through human-computer interaction. That is, each frame of contrast imaging image is displayed on the computer device's screen; the device receives user trigger operations on each frame of contrast imaging image and determines the initial ROI and initial non-ROI in each frame of contrast imaging image.

[0149] In another alternative embodiment, the computer device inputs the contrast-enhanced video into a pre-trained recognition model, and through the recognition processing of the recognition model, obtains the initial region of interest and the initial region of non-interest in each frame of the contrast-enhanced image.

[0150] Step 910: Based on the motion tracking algorithm, correct the initial region of interest and the initial region of non-interest to obtain the target region of interest and the target region of non-interest.

[0151] Motion tracking algorithms can include frame differencing, segmentation, and kernel-based correlation filters (KCF). After obtaining the initial region of interest (ROI) and initial non-ROI in each frame of contrast imaging, the computer device tracks these ROIs based on motion tracking algorithms. This allows for the correction of the ROIs in each frame to obtain the target ROI and target non-ROI. Different motion tracking algorithms result in different specific processes for correcting the ROIs in each frame; this embodiment does not impose limitations on this, as long as the functionality is achieved.

[0152] In this embodiment, the initial region of interest (ROI) and initial non-ROI in each frame of the contrast image are corrected based on a motion tracking algorithm to obtain the target ROI and target non-ROI. This avoids the influence of the motion of the object to be processed on the determined ROI and target ROI, thereby improving the accuracy of the determined target ROI and target non-ROI.

[0153] In an optional embodiment, after determining the keyframe images, the method further includes displaying each keyframe image on a clipboard and storing it in the memory of a computer device. The computer device may also display each keyframe image on a display screen for the doctor to view, in order to output a corresponding examination report.

[0154] Please see Figure 10 One embodiment of this application provides an image processing method, the steps of which include:

[0155] Step 100: Acquire contrast imaging video; determine the initial region of interest and initial region of non-interest corresponding to each frame of contrast imaging image; the contrast imaging video includes multiple frames of contrast imaging images;

[0156] Step 101: Based on the motion tracking algorithm, correct the initial region of interest and the initial region of non-interest to obtain the target region of interest and the target region of non-interest;

[0157] Step 102: For each frame of the imaging image, determine the first mean of the pixel values ​​of all pixels in the target region of interest, and determine the second mean of the pixel values ​​of all pixels in the target region of non-interest.

[0158] Step 103: Determine the first imaging intensity value corresponding to the target region of interest based on the first mean, and determine the second imaging intensity value corresponding to the target region of non-interest based on the second mean.

[0159] Step 104: Perform fitting analysis on the first contrast intensity value of all frames to obtain the first variation curve, and perform fitting analysis on the second contrast intensity value of all frames to obtain the second variation curve.

[0160] Step 105: Determine the first keyframe image, the second keyframe image, and the third keyframe image based on the starting point, peak point, and rising point of the first change curve.

[0161] Step 106: Determine the fourth keyframe image based on the intersection point between the first descending curve segment of the first change curve and the second descending curve segment of the second change curve.

[0162] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0163] Based on the same inventive concept, this application also provides an image processing apparatus for implementing the image processing method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more image processing apparatus embodiments provided below can be found in the limitations of the image processing method described above, and will not be repeated here.

[0164] In one embodiment, such as Figure 11 As shown, an image processing apparatus 10 is provided, including: an acquisition module 11, an information determination module 12, and an image determination module 13, wherein:

[0165] Acquisition module 11 is used to acquire contrast imaging video; the contrast imaging video includes multiple frames of contrast imaging images;

[0166] The information determination module 12 is used to determine the target region of interest and the target region of non-interest corresponding to each frame of the imaging image, and to determine the first feature change information of the target region of interest and the second feature change information of the target region of non-interest;

[0167] The image determination module 13 is used to determine the key frame image based on the first feature change information and the second feature change information.

[0168] In one embodiment, the image determination module 13 includes a first image determination unit and a second image determination unit. The first image determination unit is used to determine a first keyframe image, a second keyframe image, and a third keyframe image based on first feature change information. The second image determination unit is used to determine a fourth keyframe image based on the first feature change information and the second feature change information.

[0169] In one embodiment, the first determining unit is specifically configured to determine a first keyframe image based on the starting point of the first change curve corresponding to the first feature change information; the first determining unit is further configured to determine a second keyframe image based on the peak point of the first change curve corresponding to the first feature change information; the first determining unit is further configured to determine the rising point of the first change curve based on the starting point of the first change curve corresponding to the first feature change information; and determine a third keyframe image based on the rising point of the first change curve.

[0170] In one embodiment, the first determining unit is further configured to determine the frame images corresponding to the average value of the starting points of all the first change curves as the first keyframe images; the first determining unit is further configured to determine the frame images corresponding to the average value of the peak points of all the first change curves as the second keyframe images; and the first determining unit is further configured to determine the frame images corresponding to the average value of the rising points of all the first change curves as the third keyframe images.

[0171] In one embodiment, the second determining unit is specifically used to determine the first descending curve segment in the first change curve corresponding to the first feature change information; determine the second descending curve segment in the second change curve corresponding to the second feature change information; and determine the fourth keyframe image based on the intersection point between the first descending curve segment and the second descending curve segment.

[0172] In one embodiment, the second determining unit is further configured to determine the frame image corresponding to the average value of the intersection points between all the first descending curve segments and the second descending curve segments as the fourth keyframe image.

[0173] In one embodiment, the information determination module 12 includes an average determination unit and an information determination unit. The average determination unit is used to determine, for each frame of the imaging image, a first average of the pixel values ​​of all pixels in the target region of interest, and a second average of the pixel values ​​of all pixels in the target region of non-interest. The information determination unit determines first feature change information based on the first average of all frames, and determines second feature change information based on the second average of all frames.

[0174] In one embodiment, the information determining unit is specifically configured to determine a first contrast intensity value corresponding to the target region of interest in all frames based on a first mean of all frames, and to determine a second contrast intensity value corresponding to the target region of non-interest in all frames based on a second mean of all frames; determine first feature change information based on the first contrast intensity value of all frames, and determine second feature change information based on the second contrast intensity value of all frames.

[0175] In one embodiment, the information determining device 12 further includes a correction unit. The correction unit is used to determine the initial region of interest and the initial region of non-interest corresponding to each frame of the imaging image; based on a motion tracking algorithm, the initial region of interest and the initial region of non-interest are corrected to obtain the target region of interest and the target region of non-interest.

[0176] Each module in the aforementioned image processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0177] In one embodiment, a computer device is provided, the internal structure of which can be shown as follows: Figure 1 As shown. Those skilled in the art will understand that... Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0178] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0179] Acquire contrast imaging video; the contrast imaging video includes multiple frames of contrast imaging images;

[0180] The target region of interest and the target region of non-interest are determined for each frame of the imaging image, and the first feature change information of the target region of interest and the second feature change information of the target region of non-interest are determined.

[0181] The keyframe image is determined based on the first feature change information and the second feature change information.

[0182] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0183] Acquire contrast imaging video; the contrast imaging video includes multiple frames of contrast imaging images;

[0184] The target region of interest and the target region of non-interest are determined for each frame of the imaging image, and the first feature change information of the target region of interest and the second feature change information of the target region of non-interest are determined.

[0185] The keyframe image is determined based on the first feature change information and the second feature change information.

[0186] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0187] Acquire contrast imaging video; the contrast imaging video includes multiple frames of contrast imaging images;

[0188] The target region of interest and the target region of non-interest are determined for each frame of the imaging image, and the first feature change information of the target region of interest and the second feature change information of the target region of non-interest are determined.

[0189] The keyframe image is determined based on the first feature change information and the second feature change information.

[0190] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0191] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above 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.

[0192] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An image processing method, characterized in that, The method includes: Acquire contrast-enhanced video; the contrast-enhanced video includes multiple frames of contrast-enhanced images; The target region of interest and the target region of non-interest corresponding to each frame of the imaging image are determined, and the first feature change information of the target region of interest and the second feature change information of the target region of non-interest are determined; Based on the first feature change information and the second feature change information, a key frame image is determined; the key frame image includes a fourth key frame image. Determining the keyframe image based on the first feature change information and the second feature change information includes: Determine the first descending curve segment of the first change curve corresponding to the first feature change information; determine the second descending curve segment of the second change curve corresponding to the second feature change information; When there are multiple target regions of interest and multiple corresponding first change curves, the frame image corresponding to the average value of the intersection points between all the first and second descending curve segments is determined as the fourth keyframe image.

2. The method according to claim 1, characterized in that, The keyframe image further includes a first keyframe image, a second keyframe image, and a third keyframe image. The step of determining the keyframe image based on the first feature change information and the second feature change information includes: Based on the first feature change information, the first keyframe image, the second keyframe image, and the third keyframe image are determined.

3. The method according to claim 2, characterized in that, The step of determining the first keyframe image, the second keyframe image, and the third keyframe image based on the first feature change information includes: The first keyframe image is determined based on the starting point of the first change curve corresponding to the first feature change information. The second keyframe image is determined based on the peak point of the first change curve corresponding to the first feature change information; Based on the starting point of the first change curve corresponding to the first feature change information, the rising point of the first change curve is determined; and based on the rising point of the first change curve, the third keyframe image is determined.

4. The method according to claim 3, characterized in that, The target region of interest is multiple, and the corresponding first change curves are multiple. Determining the first keyframe image, the second keyframe image, and the third keyframe image based on the first feature change information includes: The frame image corresponding to the average value of the starting points of all the first change curves is determined as the first keyframe image; The frame image corresponding to the average value of all peak points of the first change curve is determined as the second key frame image; The frame image corresponding to the average value of the rising points of all the first change curves is determined as the third keyframe image.

5. The method according to any one of claims 1-4, characterized in that, The determination of the first feature change information of the target region of interest and the second feature change information of the target region of non-interest includes: For each frame of the imaging image, a first mean of the pixel values ​​of all pixels in the target region of interest is determined, and a second mean of the pixel values ​​of all pixels in the target region of non-interest is determined. The first feature change information is determined based on the first mean of all frames, and the second feature change information is determined based on the second mean of all frames.

6. The method according to claim 5, characterized in that, The step of determining the first feature change information based on the first mean of all frames, and determining the second feature change information based on the second mean of all frames, includes: Based on the first mean of all frames, a first contrast intensity value corresponding to the target region of interest in all frames is determined, and based on the second mean of all frames, a second contrast intensity value corresponding to the target region of non-interest in all frames is determined. The first feature change information is determined based on the first contrast intensity value of all frames, and the second feature change information is determined based on the second contrast intensity value of all frames.

7. The method according to any one of claims 1-4, characterized in that, Determining the target region of interest and target region of non-interest corresponding to each frame of the imaging image includes: Determine the initial region of interest and the initial region of non-interest corresponding to each frame of the imaging image; Based on a motion tracking algorithm, the initial region of interest and the initial region of non-interest are corrected to obtain the target region of interest and the target region of non-interest.

8. An image processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire contrast imaging video; the contrast imaging video includes multiple frames of contrast imaging images; The information determination module is used to determine the target region of interest and the target region of non-interest corresponding to each frame of the imaging image, and to determine the first feature change information of the target region of interest and the second feature change information of the target region of non-interest; An image determination module is used to determine a keyframe image based on the first feature change information and the second feature change information; the keyframe image includes a fourth keyframe image; The image determination module includes a second determination unit, which is used to determine the first descending curve segment of the first change curve corresponding to the first feature change information; determine the second descending curve segment of the second change curve corresponding to the second feature change information; and, when there are multiple target regions of interest and multiple corresponding first change curves, determine the frame image corresponding to the average value of all intersection points between the first descending curve segments and the second descending curve segments as the fourth keyframe image.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images