Image processing method and apparatus, program product, imaging device, and surgical robot

Abstract

The present application provides an image processing method and apparatus, a program product, an imaging device, and a surgical robot. The image processing method comprises: acquiring a three-dimensional image of a first subject under detection, and converting the three-dimensional image into a plurality of two-dimensional images; and performing contour extraction on the plurality of two-dimensional images to obtain a first contour image of said first subject.

Description

Image processing method, device, program product, imaging device and surgical robot

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on July 17, 2023, with application number "202310877470.5" and application name "Image processing method, device, program product, imaging equipment and surgical robot", the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of image processing technology, and in particular to an image processing method, apparatus, program product, imaging equipment, and surgical robot. Background Art

[0003] Currently, segmenting the skin and bed surface in CT (Computered Tomography) chest images is a key step for surgical robots to perform precise operations. However, due to the limited number and quality of public training sets, traditional image processing methods are often used for this purpose.

[0004] Application Contents

[0005] The present application aims to at least solve the problems of low image accuracy and relatively low image robustness existing in the prior art or related technologies.

[0006] To this end, a first aspect of the present application is to propose an image processing method.

[0007] The second aspect of this application is to propose another image processing method.

[0008] The third aspect of the present application is to provide an image processing device.

[0009] The fourth aspect of the present application is to provide another image processing device.

[0010] A fifth aspect of the present application is to provide a readable storage medium.

[0011] A sixth aspect of the present application is to provide a computer program product.

[0012] A seventh aspect of the present application is to provide a computer device.

[0013] An eighth aspect of the present application is to provide an imaging device.

[0014] The ninth aspect of this application is to propose a surgical robot.

[0015] In view of this, according to the first aspect of the present application, an image processing method is proposed, which includes: obtaining a three-dimensional image of a first detection object, converting the three-dimensional image into multiple two-dimensional images; and performing contour extraction on the multiple two-dimensional images to obtain a first contour image of the first detection object.

[0016] The image processing method in this technical solution improves the accuracy of the contour information in the first contour image, while improving the precision and robustness of the first contour image.

[0017] According to the second aspect of the present application, another image processing method is proposed, which includes: obtaining a three-dimensional image including a first detection object and a second detection object, the second detection object being used to carry the first detection object, and determining a first contour image of the first detection object based on the three-dimensional image; and performing feature extraction on the three-dimensional image based on the first contour image to obtain a second contour image of the second detection object.

[0018] The image processing method in this technical solution improves the accuracy of the contour information in the second contour image, while improving the precision and robustness of the second contour image.

[0019] According to the third aspect of the present application, an image processing device is proposed, which includes: a first processing module, used to obtain a three-dimensional image of a first detection object and convert the three-dimensional image into multiple two-dimensional images; the first processing module is also used to extract contours from the multiple two-dimensional images to obtain a first contour image of the first detection object.

[0020] The image processing device in this technical solution improves the accuracy of the contour information in the first contour image, while improving the precision and robustness of the first contour image.

[0021] According to the fourth aspect of the present application, another image processing device is proposed, which includes: a second processing module for obtaining a three-dimensional image including a first detection object and a second detection object, the second detection object being used to carry the first detection object, and determining a first contour image of the first detection object based on the three-dimensional image; the second processing module is also used to extract features from the three-dimensional image based on the first contour image to obtain a second contour image of the second detection object.

[0022] The image processing device in this technical solution improves the accuracy of the contour information in the second contour image, while improving the precision and robustness of the second contour image.

[0023] According to a fifth aspect of the present application, a readable storage medium is provided, on which a program or instruction is stored. When executed by a processor, the program or instruction implements the image processing method described in any of the above technical solutions. Therefore, the readable storage medium has all the beneficial effects of the image processing method described in any of the above technical solutions, and no further description is given here.

[0024] According to a sixth aspect of the present application, a computer program product is provided, comprising computer instructions. When executed by a processor, the computer instructions implement the image processing method described in any of the above technical solutions. Therefore, the readable storage medium has all the beneficial effects of the image processing method described in any of the above technical solutions, and no further description is given here.

[0025] According to a seventh aspect of the present application, a computer device is provided, comprising a processor and a memory, wherein the memory stores a program or instructions, which, when executed by the processor, implement the steps of the image processing method described in any of the above technical solutions. Therefore, the computer device possesses all the beneficial effects of the image processing method described in any of the above technical solutions, and further description thereof is omitted here.

[0026] According to the eighth aspect of the present application, an imaging device is proposed, comprising: an image processing device as defined in the third aspect above, or the image processing device defined in the fourth aspect above, and / or the readable storage medium defined in the fifth aspect above, and / or the computer program product defined in the sixth aspect above, thereby having all the beneficial technical effects of the image processing device defined in the third aspect above, or the image processing device defined in the fourth aspect above, and / or the readable storage medium defined in the fifth aspect above, and / or the computer program product defined in the sixth aspect above, and no further details will be given here.

[0027] According to the ninth aspect of the present application, a surgical robot is proposed, which plans a motion trajectory or avoids obstacles based on a processed image, wherein the image is obtained based on the steps of the image processing method defined in the first aspect or the image processing method defined in the second aspect, and thus has all the beneficial technical effects of the image processing method defined in the first aspect or the image processing method defined in the second aspect, and will not be elaborated on here.

[0028] Additional aspects and advantages of the present application will become apparent in the following description or may be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0030] FIG1 shows one of the flow charts of the image processing method in an embodiment of the present application;

[0031] FIG2 shows a second flow chart of the image processing method in an embodiment of the present application;

[0032] FIG3 shows a third flow chart of the image processing method in an embodiment of the present application;

[0033] FIG4 shows a fourth flow chart of the image processing method in an embodiment of the present application;

[0034] FIG5 shows a fifth flow chart of the image processing method in an embodiment of the present application;

[0035] FIG6 shows a sixth flow chart of the image processing method in an embodiment of the present application;

[0036] FIG7 shows a seventh flow chart of the image processing method in an embodiment of the present application;

[0037] FIG8 shows an eighth flowchart of the image processing method in an embodiment of the present application;

[0038] FIG9 shows a ninth flowchart of the image processing method in an embodiment of the present application;

[0039] FIG10 shows a tenth flowchart of the image processing method in an embodiment of the present application;

[0040] FIG11 shows an eleventh flowchart of the image processing method in an embodiment of the present application;

[0041] FIG12 shows a twelfth flowchart of the image processing method in an embodiment of the present application;

[0042] FIG13 shows one structural block diagram of an image processing apparatus in an embodiment of the present application;

[0043] FIG14 shows one of the schematic diagrams of the image processing device in the embodiment of the present application;

[0044] FIG15 shows a second schematic diagram of the image processing device in an embodiment of the present application;

[0045] FIG16 shows a third schematic diagram of the image processing device in an embodiment of the present application;

[0046] FIG17 shows a second structural block diagram of the image processing device in an embodiment of the present application;

[0047] FIG18 shows a fourth schematic diagram of the image processing device in an embodiment of the present application;

[0048] FIG19 shows a fifth schematic diagram of the image processing device in an embodiment of the present application;

[0049] FIG20 shows a sixth schematic diagram of the image processing device in an embodiment of the present application;

[0050] FIG21 shows a third structural block diagram of the image processing device in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to more clearly understand the above-mentioned objects, features and advantages of the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features therein can be combined with each other in the absence of conflict.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present application is not limited to the specific embodiments disclosed below.

[0053] The image processing method, apparatus, program product, imaging device, and surgical robot provided in the embodiments of the present application are described in detail below with reference to Figures 1 to 21 through specific embodiments and their application scenarios.

[0054] The image processing method provided in this application may be implemented by an image processing device, which may also be determined based on actual usage requirements and is not specifically limited herein. To more clearly describe the image processing method provided in this application, the following description will be based on an image processing device as the implementing body.

[0055] In one embodiment of the present application, as shown in FIG1 , an image processing method is proposed. The image processing method includes:

[0056] Step 102: Acquire a three-dimensional image of a first detection object and convert the three-dimensional image into a plurality of two-dimensional images;

[0057] Step 104 : performing contour extraction on the multiple two-dimensional images to obtain a first contour image of the first detection object.

[0058] In this embodiment, an image processing method is provided. The image processing method is used to segment contour information of a first detection object, wherein the first detection object is an object to be detected.

[0059] Exemplarily, the first detection object may be specifically the chest cavity of a patient.

[0060] The image processing device obtains a three-dimensional image of the first detection object and performs dimension conversion on the three-dimensional image to obtain multiple two-dimensional images, wherein the three-dimensional image is a three-dimensional image containing the first detection object and the two-dimensional image is a two-dimensional image containing the first detection object.

[0061] For example, the three-dimensional image may specifically be a three-dimensional CT image of the first detection object acquired by a CT device.

[0062] The image processing device performs contour segmentation processing on the multiple two-dimensional images respectively to determine a first contour image of the first detection object.

[0063] For example, the first contour image may be a patient's chest skin contour image.

[0064] Exemplarily, the image processing device performs threshold segmentation on the input three-dimensional image. Then, applying the prior knowledge that the skin layer contour in the axial position of the chest image is the largest, the device extracts the maximum contour layer by layer, removes excess tissue in the non-contour area, and completes the skin segmentation. It should be noted that due to differences in the CT value range of the skin in CT images from different manufacturers, models, or imaging parameters, there is no standard set of thresholds that can accurately segment the skin. Since this embodiment will subsequently perform contour extraction on the threshold segmentation results, the accuracy requirements for the set thresholds are reduced. Therefore, the range of upper and lower threshold limits that can be set is relatively large, reducing the reliance on the threshold.

[0065] The image processing method in this embodiment obtains a three-dimensional image of the first detection object, performs dimensionality conversion on the three-dimensional image to obtain multiple two-dimensional images, and then performs contour segmentation processing on the multiple two-dimensional images to determine the first contour image of the first detection object, thereby improving the accuracy of the contour information in the first contour image, and at the same time improving the precision and robustness of the first contour image.

[0066] In one embodiment of the present application, as shown in FIG2 , an image processing method is proposed. The image processing method includes:

[0067] Step 202: Acquire a three-dimensional image of a first detection object and convert the three-dimensional image into a plurality of two-dimensional images;

[0068] Step 204: determining target contour information of the first detection object through the multiple two-dimensional images;

[0069] Step 206 : Obtain a range threshold of the three-dimensional image, and perform image segmentation processing on the multiple two-dimensional images according to the range threshold and the target contour information to obtain a first contour image.

[0070] In this embodiment, the image processing device extracts contour features from a plurality of two-dimensional images to obtain target contour information of the first detection object, wherein the target contour information is distribution information of the contour of the first detection object.

[0071] For example, the target contour information may specifically be distribution information of the skin contour.

[0072] The image processing device obtains a range threshold of a three-dimensional image, and performs image segmentation on multiple two-dimensional images according to the range threshold and target contour information, thereby obtaining a first contour image of the first detection object, wherein the range threshold is the threshold for image segmentation.

[0073] Exemplarily, the range threshold may be specifically [-200, 100].

[0074] For example, the plurality of two-dimensional images may include a plurality of layer images along the z-axis direction.

[0075] Exemplarily, the image processing apparatus may retrieve the maximum contour in the current layer layer by layer along the z-axis direction of the three-dimensional image and delete the remaining segmentation results.

[0076] The image processing method in this embodiment extracts contour features from multiple two-dimensional images to obtain target contour information of the first detection object, and then performs image segmentation on the multiple two-dimensional images layer by layer based on the range threshold and the target contour information to obtain a first contour image of the first detection object, thereby ensuring the information accuracy of the target contour information and thus ensuring the image accuracy of the first contour image.

[0077] In one embodiment of the present application, as shown in FIG3 , an image processing method is proposed. The image processing method includes:

[0078] Step 302: Acquire a three-dimensional image of a first detection object and convert the three-dimensional image into a plurality of two-dimensional images;

[0079] Step 304: performing contour detection on the plurality of two-dimensional images to obtain a plurality of contour information;

[0080] Step 306: performing data fusion on the multiple contour information to obtain target contour information from the multiple contour information;

[0081] Step 308 : Obtain a range threshold of the three-dimensional image, and perform image segmentation processing on the multiple two-dimensional images according to the range threshold and the target contour information to obtain a first contour image.

[0082] In this embodiment, the image processing device performs contour detection on a plurality of two-dimensional images respectively to determine a plurality of contour information, wherein the contour information is distribution information of the contours in the two-dimensional images.

[0083] For example, the contour information may specifically be distribution information of the skin in a two-dimensional image.

[0084] The image processing device performs data fusion on the multiple contour information and converts the multiple contour information into target contour information.

[0085] For example, the image processing device may perform data comparison on a plurality of contour information, and then select the target contour information from the plurality of contour information.

[0086] The image processing method in this embodiment performs contour detection on multiple two-dimensional images to determine multiple contour information, and then performs data fusion on the multiple contour information to convert the multiple contour information into target contour information, thereby ensuring the information accuracy of the target contour information and further ensuring the image accuracy of the first contour image.

[0087] In one embodiment of the present application, as shown in FIG4 , an image processing method is proposed. The image processing method includes:

[0088] Step 402: Acquire a three-dimensional image of the first detection object, obtain a target pixel axis of the three-dimensional image, and perform image conversion processing on the three-dimensional image according to the target pixel axis to obtain a plurality of two-dimensional images;

[0089] Step 404 : performing contour extraction on the multiple two-dimensional images to obtain a first contour image of the first detection object.

[0090] In this embodiment, the image processing device obtains a target pixel axis of a three-dimensional image, and then converts the three-dimensional image into multiple two-dimensional images along the target pixel axis, wherein the target pixel axis is a pixel axis in the three-dimensional image.

[0091] Exemplarily, the target pixel axis may be specifically an X-axis, a Y-axis, or a Z-axis.

[0092] The image processing method in this embodiment converts a three-dimensional image into multiple two-dimensional images through a target pixel axis, thereby ensuring the image accuracy of the multiple two-dimensional images and further ensuring the information accuracy of the first contour image.

[0093] In one embodiment of the present application, as shown in FIG5 , an image processing method is proposed. The image processing method includes:

[0094] Step 502: Acquire an original image of the first detection object, perform denoising and enhancement processing on the original image to obtain a three-dimensional image, and convert the three-dimensional image into multiple two-dimensional images;

[0095] Step 504 : performing contour extraction on the multiple two-dimensional images to obtain a first contour image of the first detection object.

[0096] In this embodiment, the image processing device obtains an original image of the first detection object, wherein the original image is an initial image of the first detection object.

[0097] Exemplarily, the original image may specifically be an initial CT image of the first detection object.

[0098] The image processing device performs image preprocessing such as denoising and enhancement on the original image to obtain a three-dimensional image.

[0099] Exemplarily, the image processing device may perform pre-processing such as noise reduction on the original image to obtain a three-dimensional image.

[0100] The image processing method in this embodiment obtains an original image of the first detection object and performs image preprocessing such as denoising and enhancement on the original image to obtain a three-dimensional image, thereby ensuring the image accuracy of the three-dimensional image and further ensuring the image accuracy of the first contour image.

[0101] In one embodiment of the present application, as shown in FIG6 , an image processing method is proposed. The image processing method includes:

[0102] Step 602: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object based on the three-dimensional image;

[0103] Step 604 : performing feature extraction on the three-dimensional image based on the first contour image to obtain a second contour image of the second detection object.

[0104] In this embodiment, an image processing method is provided. Specifically, an image processing device obtains a three-dimensional image, wherein the three-dimensional image is a three-dimensional image including a first detection object and a second detection object, and the second detection object is used to carry the first detection object.

[0105] For example, the first detection object may be specifically the chest cavity of a patient, or the first detection object may be specifically the bed board supporting the patient.

[0106] The image processing device performs image segmentation on the three-dimensional image to obtain a first contour image of the first detection object, and then performs image segmentation on the three-dimensional image based on the first contour image to obtain a second contour image of the second detection object, wherein the second contour image is a distribution image of the contour of the second detection object.

[0107] For example, the second contour image may be a contour image of a bed board in a three-dimensional image, where the bed board is a bed board in a CT imaging device.

[0108] The image processing method in this embodiment acquires a three-dimensional image, performs image segmentation on the three-dimensional image, and obtains a first contour image of the first detection object. Then, based on the first contour image, the three-dimensional image is segmented to obtain a second contour image of the second detection object, thereby improving the accuracy of the contour information in the second contour image and improving the precision and robustness of the second contour image.

[0109] In one embodiment of the present application, as shown in FIG7 , an image processing method is proposed. The image processing method includes:

[0110] Step 702: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object based on the three-dimensional image;

[0111] Step 704, determining non-background pixels in the three-dimensional image based on the first contour image;

[0112] Step 706: Obtain a threshold combination, and determine a target threshold in the threshold combination based on non-background pixels;

[0113] Step 708: Perform image processing on the three-dimensional image according to the target threshold to determine a second contour image.

[0114] In this embodiment, the image processing device determines non-background pixels in the three-dimensional image based on the first contour image, wherein the non-background pixels are pixels in the non-background part of the three-dimensional image.

[0115] For example, the non-background pixels in the three-dimensional image may specifically be pixel points in a non-bedboard portion of the three-dimensional image.

[0116] The image processing device obtains a preset threshold combination and determines a target threshold in the threshold combination based on non-background pixels, wherein the threshold combination is a preset threshold combination, including thresholds corresponding to second detection objects of different materials, and the target threshold is the threshold in the threshold combination.

[0117] Exemplarily, the threshold combination includes thresholds corresponding to bed boards of different materials.

[0118] For example, the threshold combination can adaptively delete or add thresholds within the combination according to demand, thereby improving the application flexibility of the threshold combination and further expanding the application scope of the threshold combination.

[0119] The image processing method in this embodiment determines the non-background pixels in the three-dimensional image based on the first contour image, and then determines the target threshold in the threshold combination based on the non-background pixels, thereby ensuring the accuracy of the target threshold and further ensuring the accuracy and robustness of the second contour image.

[0120] In one embodiment of the present application, as shown in FIG8 , an image processing method is proposed. The image processing method includes:

[0121] Step 802: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object based on the three-dimensional image;

[0122] Step 804, determining non-background pixels in the three-dimensional image based on the first contour image;

[0123] Step 806, performing data processing on the non-background pixels to obtain a distribution mean of the non-background pixels;

[0124] Step 808, determining a target threshold value in the threshold value combination by performing a difference operation on the distribution mean and the threshold value combination;

[0125] Step 810: Perform image processing on the three-dimensional image according to the target threshold to determine a second contour image.

[0126] In this embodiment, the image processing apparatus performs a mean operation on the non-background pixels to obtain a distribution mean of the non-background pixels, wherein the distribution mean is the mean of the distribution information of the non-background pixels.

[0127] Exemplarily, the distribution mean may specifically be the mean of the coordinate values ​​of non-background pixels.

[0128] The image processing device performs a difference operation on the distribution mean and the threshold combination to determine a target threshold in the threshold combination.

[0129] Exemplarily, the image processing apparatus performs a difference operation on the distribution mean and the threshold combination, and determines a target threshold in the threshold combination according to a minimum difference.

[0130] The image processing method in this embodiment performs a mean operation on non-background pixels to obtain the distribution mean of the non-background pixels, and performs a difference operation on the distribution mean and the threshold combination to determine the target threshold in the threshold combination, thereby ensuring the accuracy of the target threshold and further ensuring the data accuracy of the second contour image.

[0131] In one embodiment of the present application, as shown in FIG9 , an image processing method is proposed. The image processing method includes:

[0132] Step 902: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object based on the three-dimensional image;

[0133] Step 904, determining non-background pixels in the three-dimensional image based on the first contour image;

[0134] Step 906: Obtain a threshold combination, and determine a target threshold in the threshold combination based on non-background pixels;

[0135] Step 908: performing image segmentation on the three-dimensional image according to the target threshold to obtain a segmentation result image;

[0136] Step 910: Perform noise filtering on the segmentation result image to determine a second contour image.

[0137] In this embodiment, the image processing device performs image segmentation processing on the three-dimensional image according to the target threshold value to obtain a segmentation result image, wherein the segmentation result image is an image after the three-dimensional image is segmented.

[0138] For example, the segmentation result image may specifically be a bed board image segmented from the three-dimensional image.

[0139] The image processing device performs noise filtering and other processing on the segmentation result image to obtain a second contour image.

[0140] Exemplarily, the image processing device may perform denoising and enhancement processing on the segmentation result image to obtain a second contour image.

[0141] The image processing method in this embodiment performs image segmentation processing on the three-dimensional image according to the target threshold to obtain a segmentation result image, and performs noise filtering and other processing on the segmentation result image to obtain a second contour image, thereby ensuring the image quality of the second contour image.

[0142] In one embodiment of the present application, as shown in FIG10 , an image processing method is proposed. The image processing method includes:

[0143] Step 1002: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object based on the three-dimensional image;

[0144] Step 1004: determining a pixel set of a first detection object based on the first contour image;

[0145] Step 1006, determining background pixels in the three-dimensional image based on the pixel set;

[0146] Step 1008: Classify the pixels in the three-dimensional image according to the background pixels to obtain non-background pixels;

[0147] Step 1010: Obtain a threshold combination, and determine a target threshold in the threshold combination based on non-background pixels;

[0148] Step 1012: Perform image processing on the three-dimensional image according to the target threshold to determine a second contour image.

[0149] In this embodiment, the image processing device determines a pixel set corresponding to the first detection object based on the first contour image, wherein the pixel set is the pixels of the first detection object in the three-dimensional image.

[0150] Exemplarily, the first detection object may be specifically a chest cavity, and the pixel set may include the chest cavity and pixel points inside the chest cavity.

[0151] The image processing device sets a pixel set as background pixels in a three-dimensional image, wherein the background pixels are pixels in a background portion of the three-dimensional image.

[0152] Exemplarily, the image processing apparatus may set all pixel sets corresponding to the first detection object as background pixels in the three-dimensional image.

[0153] The image processing device classifies pixels in the three-dimensional image according to background pixels to obtain non-background pixels in the three-dimensional image.

[0154] Exemplarily, pixels in a three-dimensional image can be divided into two categories: non-background pixels and background pixels.

[0155] The image processing method in this embodiment determines a pixel set corresponding to the first detection object based on the first contour image, sets the pixel set as background pixels in the three-dimensional image, and classifies the pixels in the three-dimensional image based on the background pixels to obtain non-background pixels in the three-dimensional image, thereby improving the recognition accuracy of non-background pixels in the three-dimensional image.

[0156] In one embodiment of the present application, as shown in FIG11 , an image processing method is proposed. The image processing method includes:

[0157] Step 1102: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and convert the three-dimensional image into multiple two-dimensional images;

[0158] Step 1104: performing contour extraction on the multiple two-dimensional images to obtain a first contour image of the first detection object;

[0159] Step 1106 : performing feature extraction on the three-dimensional image based on the first contour image to obtain a second contour image of the second detection object.

[0160] In this embodiment, the image processing device performs dimension conversion on the three-dimensional image to obtain multiple two-dimensional images, wherein the three-dimensional image is a three-dimensional image containing the first detection object, and the two-dimensional image is a two-dimensional image containing the first detection object.

[0161] For example, the three-dimensional image may specifically be a three-dimensional CT image of the first detection object acquired by a CT device.

[0162] The image processing device performs contour segmentation processing on the multiple two-dimensional images respectively to determine a first contour image of the first detection object.

[0163] For example, the first contour image may be a patient's chest skin contour image.

[0164] The image processing method in this embodiment obtains a three-dimensional image of the first detection object, performs dimensionality conversion on the three-dimensional image to obtain multiple two-dimensional images, and then performs contour segmentation processing on the multiple two-dimensional images to determine the first contour image of the first detection object, thereby improving the accuracy of the contour information in the first contour image, and at the same time improving the precision and robustness of the first contour image.

[0165] In one embodiment of the present application, as shown in FIG12 , an image processing method is proposed. The image processing method includes:

[0166] Step 1202: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and convert the three-dimensional image into multiple two-dimensional images.

[0167] Step 1204: determining target contour information of the first detection object based on the multiple two-dimensional images;

[0168] Step 1206: Obtain range thresholds of the plurality of two-dimensional images, and perform image segmentation processing on the plurality of two-dimensional images according to the range thresholds and target contour information to obtain a first contour image;

[0169] Step 1208 : performing feature extraction on the three-dimensional image based on the first contour image to obtain a second contour image of the second detection object.

[0170] In this embodiment, the image processing device extracts contour features from a plurality of two-dimensional images to obtain target contour information of the first detection object, wherein the target contour information is distribution information of the contour of the first detection object.

[0171] For example, the target contour information may specifically be distribution information of the skin contour.

[0172] The image processing device obtains a range threshold of multiple two-dimensional images, and performs image segmentation on the multiple two-dimensional images according to the range threshold and target contour information, thereby obtaining a first contour image of the first detection object, wherein the range threshold is the threshold for image segmentation.

[0173] The image processing method in this embodiment extracts contour features from multiple two-dimensional images to obtain target contour information of the first detection object, and then performs image segmentation on the multiple two-dimensional images based on the range threshold and the target contour information to obtain a first contour image of the first detection object, thereby ensuring the information accuracy of the target contour information and thus ensuring the image accuracy of the first contour image.

[0174] As shown in FIG13 , an embodiment of the present application provides an image processing device, wherein the image processing device 1300 includes:

[0175] A first processing module 1302 is configured to obtain a three-dimensional image of a first detection object and convert the three-dimensional image into a plurality of two-dimensional images;

[0176] The first processing module 1302 is further configured to perform contour extraction on the multiple two-dimensional images to obtain a first contour image of the first detection object.

[0177] In this embodiment, an image processing device 1300 is provided for segmenting contour information of a first detection object, wherein the first detection object is an object to be detected.

[0178] Exemplarily, the first detection object may be specifically the chest cavity of a patient.

[0179] The first processing module 1302 obtains a three-dimensional image of the first detection object and performs dimension conversion on the three-dimensional image to obtain multiple two-dimensional images, wherein the three-dimensional image is a three-dimensional image containing the first detection object, and the two-dimensional image is a two-dimensional image containing the first detection object.

[0180] For example, the three-dimensional image may specifically be a three-dimensional CT image of the first detection object acquired by a CT device.

[0181] The first processing module 1302 performs contour segmentation processing on the multiple two-dimensional images respectively to determine a first contour image of the first detection object.

[0182] For example, the first contour image may be a patient's chest skin contour image.

[0183] Exemplarily, the image processing device performs threshold segmentation on the input three-dimensional image. Then, applying the prior knowledge that the skin layer contour in the axial position of the chest image is the largest, the device extracts the maximum contour layer by layer, removes excess tissue in the non-contour area, and completes the skin segmentation. It should be noted that due to differences in the CT value range of skin in CT images from different manufacturers, models, or imaging parameters, there is no standard set of thresholds that can accurately segment the skin. Given that this embodiment will subsequently perform contour extraction on the threshold segmentation results, the accuracy requirements for the set thresholds are reduced. Therefore, the upper and lower limits of the thresholds that can be set are relatively large, reducing the reliance on the thresholds.

[0184] For example, as shown in Figure 14, a 3D CT image is input, preprocessed with denoising and enhancement, and a threshold of [-200, 100] is set for threshold segmentation (the segmentation result is shown in Figure 15). The maximum contour in the current layer is retrieved layer by layer along the z-axis, and the remaining segmentation results are deleted (the segmentation result is shown in Figure 16). As shown in Figure 15, the skin cannot be accurately segmented after threshold segmentation. By retrieving the maximum contour layer by layer in the axial slices and retaining the maximum contour, the influence of the bed board and other tissues in the body is eliminated. This process can appropriately relax the threshold requirement. Even if other tissues in the body such as lungs and bones are segmented, the skin layer can still be accurately retained after extracting the maximum contour.

[0185] It should be noted that, while conventional segmentation uses a threshold of [min, max] to accurately segment the target area, where min is the minimum value and max is the maximum value, the threshold in this embodiment can be set to [min-n, max+n], where n is a random constant. Generally, thresholds are set empirically. This embodiment does not rely heavily on the rough results of threshold segmentation and does not have a specific standard. In other words, it is not necessary to set the threshold to [min, max]. As long as the rough segmentation results include skin and are free of excessive noise, accurate segmentation can be achieved, which also improves the robustness of the algorithm. Furthermore, the skin layer in the axial plane of the chest image is the largest contour, so a looser threshold range can be set to include other tissues without any problems, as they will eventually be removed.

[0186] The image processing device 1300 in this embodiment obtains a three-dimensional image of the first detection object, performs dimensionality conversion on the three-dimensional image to obtain multiple two-dimensional images, and then performs contour segmentation processing on the multiple two-dimensional images to determine the first contour image of the first detection object, thereby improving the accuracy of the contour information in the first contour image, and at the same time improving the precision and robustness of the first contour image.

[0187] In one embodiment of the present application, the image processing apparatus 1300 further includes:

[0188] The first processing module 1302 is further configured to determine target contour information of the first detection object through the multiple two-dimensional images;

[0189] The first processing module 1302 is further configured to obtain a range threshold of the multiple two-dimensional images, and perform image segmentation processing on the multiple two-dimensional images according to the range threshold and target contour information to obtain a first contour image.

[0190] The image processing device 1300 in this embodiment extracts contour features from multiple two-dimensional images to obtain target contour information of the first detection object, and then performs image segmentation on the multiple two-dimensional images based on the range threshold and the target contour information to obtain a first contour image of the first detection object, thereby ensuring the information accuracy of the target contour information and thus ensuring the image accuracy of the first contour image.

[0191] In one embodiment of the present application, the image processing apparatus 1300 further includes:

[0192] The first processing module 1302 is further configured to perform contour detection on the plurality of two-dimensional images to obtain a plurality of contour information;

[0193] The first processing module 1302 is further configured to perform data fusion on the multiple contour information to obtain target contour information from the multiple contour information.

[0194] The image processing device 1300 in this embodiment performs contour detection on multiple two-dimensional images to determine multiple contour information, and then performs data fusion on the multiple contour information to convert the multiple contour information into target contour information, thereby ensuring the information accuracy of the target contour information and further ensuring the image accuracy of the first contour image.

[0195] In one embodiment of the present application, the image processing apparatus 1300 further includes:

[0196] The first processing module 1302 is further configured to obtain a target pixel axis of the three-dimensional image, and perform image conversion processing on the three-dimensional image according to the target pixel axis to obtain a plurality of two-dimensional images.

[0197] The image processing device 1300 in this embodiment converts a three-dimensional image into multiple two-dimensional images through the target pixel axis, thereby ensuring the image accuracy of the multiple two-dimensional images and further ensuring the information accuracy of the first contour image.

[0198] In one embodiment of the present application, the image processing apparatus 1300 further includes:

[0199] The first processing module 1302 is further configured to obtain an original image of the first detection object, and perform denoising and enhancement processing on the original image to obtain a three-dimensional image.

[0200] The image processing device 1300 in this embodiment obtains an original image of the first detection object and performs image preprocessing such as denoising and enhancement on the original image to obtain a three-dimensional image, thereby ensuring the image accuracy of the three-dimensional image and further ensuring the image accuracy of the first contour image.

[0201] As shown in FIG17 , an embodiment of the present application provides an image processing device, wherein the image processing device 1700 includes:

[0202] A second processing module 1702 is configured to obtain a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object based on the three-dimensional image;

[0203] The second processing module 1702 is further configured to perform feature extraction on the three-dimensional image according to the first contour image to obtain a second contour image of the second detection object.

[0204] In this embodiment, an image processing device 1700 is provided. Specifically, the second processing module 1702 obtains a three-dimensional image, wherein the three-dimensional image is a three-dimensional image including a first detection object and a second detection object, and the second detection object is used to carry the first detection object.

[0205] For example, the first detection object may be specifically the chest cavity of a patient, or the first detection object may be specifically the bed board supporting the patient.

[0206] The second processing module 1702 performs image segmentation on the three-dimensional image to obtain a first contour image of the first detection object, and then performs image segmentation on the three-dimensional image based on the first contour image to obtain a second contour image of the second detection object, wherein the second contour image is a distribution image of the contour of the second detection object.

[0207] For example, the second contour image may be a contour image of a bed board in a three-dimensional image, where the bed board is a bed board in a CT imaging device.

[0208] For example, as shown in Figure 18, a 3D CT image is input and preprocessed with denoising and enhancement. The CT mean of all non-background pixels in the skin segmentation results from step 3 of the threshold segmentation scheme is calculated. The absolute difference between the CT mean and the median of each threshold pair is then calculated. The group with the smallest absolute value is considered a matching threshold pair, and threshold segmentation is performed using these matching threshold pairs (the segmentation results are shown in Figure 19). The portion of the segmentation result that lies within the skin contour is removed, and isolated point noise is removed using morphological methods (the segmentation results are shown in Figure 20). Through the above steps, threshold pairs can be adaptively determined for different scanning bed materials, achieving automatic segmentation. Figure 19 shows that the threshold segmentation results contain a lot of noise other than the bed board. Because the CT value range of CT bed boards made of certain materials overlaps with the CT range of certain human tissues, such as bone, redundant tissue segmentation is inevitable after threshold segmentation. Applying the prior knowledge that the bed board and the human body are spatially separated, the area within the skin contour and the contour is definitely not the bed board area. In the coarse threshold segmentation results of the bed board, all pixels whose coordinates lie within the skin contour and the contour are set as background, reducing the dependence on the threshold value and improving the accuracy of bed board segmentation. After removing the part within the skin contour from the segmentation result and morphologically removing isolated point noise and connected domains with an area less than 300, as shown in Figure 20, the bed board is accurately segmented.

[0209] The image processing device 1700 in this embodiment acquires a three-dimensional image, performs image segmentation on the three-dimensional image, and obtains a first contour image of the first detection object. Then, based on the first contour image, the three-dimensional image is segmented to obtain a second contour image of the second detection object, thereby improving the accuracy of the contour information in the second contour image and improving the precision and robustness of the second contour image.

[0210] In one embodiment of the present application, the image processing apparatus 1700 further includes:

[0211] The second processing module 1702 is further configured to determine non-background pixels in the three-dimensional image based on the first contour image;

[0212] The second processing module 1702 is further configured to obtain a threshold combination and determine a target threshold in the threshold combination based on non-background pixels, wherein the threshold combination includes thresholds corresponding to second detection objects of different materials;

[0213] The second processing module 1702 is further configured to perform image processing on the three-dimensional image according to the target threshold value to determine a second contour image.

[0214] The image processing device 1700 in this embodiment determines the non-background pixels in the three-dimensional image based on the first contour image, and then determines the target threshold in the threshold combination based on the non-background pixels, thereby ensuring the accuracy of the target threshold and further ensuring the accuracy and robustness of the second contour image.

[0215] In one embodiment of the present application, the image processing apparatus 1700 further includes:

[0216] The second processing module 1702 is further configured to perform data processing on non-background pixels to obtain a distribution mean of the non-background pixels;

[0217] The second processing module 1702 is further configured to determine a target threshold value in the threshold value combination by performing a difference operation on the distribution mean and the threshold value combination.

[0218] The image processing device 1700 in this embodiment performs a mean operation on non-background pixels to obtain the distribution mean of the non-background pixels, and performs a difference operation on the distribution mean and the threshold combination to determine the target threshold in the threshold combination, thereby ensuring the accuracy of the target threshold and further ensuring the data accuracy of the second contour image.

[0219] In one embodiment of the present application, the image processing apparatus 1700 further includes:

[0220] The second processing module 1702 is further configured to perform image segmentation on the three-dimensional image according to the target threshold value to obtain a segmentation result image;

[0221] The second processing module 1702 is further configured to perform noise filtering on the segmentation result image to determine a second contour image.

[0222] The image processing device 1700 in this embodiment performs image segmentation processing on the three-dimensional image according to the target threshold to obtain a segmentation result image, and performs noise filtering and other processing on the segmentation result image to obtain a second contour image, thereby ensuring the image quality of the second contour image.

[0223] In one embodiment of the present application, the image processing apparatus 1700 further includes:

[0224] The second processing module 1702 is further configured to determine a pixel set of the first detection object based on the first contour image;

[0225] The second processing module 1702 is further configured to determine background pixels in the three-dimensional image based on the pixel set;

[0226] The second processing module 1702 is further configured to classify pixels in the three-dimensional image according to background pixels to obtain non-background pixels.

[0227] The image processing device 1700 in this embodiment determines a pixel set corresponding to the first detection object based on the first contour image, sets the pixel set as background pixels in the three-dimensional image, and classifies the pixels in the three-dimensional image based on the background pixels to obtain non-background pixels in the three-dimensional image, thereby improving the recognition accuracy of non-background pixels in the three-dimensional image.

[0228] In one embodiment of the present application, the image processing apparatus 1700 further includes:

[0229] The second processing module 1702 is further configured to convert the three-dimensional image into a plurality of two-dimensional images;

[0230] The second processing module 1702 is further configured to perform contour extraction on the multiple two-dimensional images to obtain a first contour image of the first detection object.

[0231] The image processing device 1700 in this embodiment obtains a three-dimensional image of the first detection object, performs dimensionality conversion on the three-dimensional image to obtain multiple two-dimensional images, and then performs contour segmentation processing on the multiple two-dimensional images to determine the first contour image of the first detection object, thereby improving the accuracy of the contour information in the first contour image, and at the same time improving the precision and robustness of the first contour image.

[0232] In one embodiment of the present application, the image processing apparatus 1700 further includes:

[0233] The second processing module 1702 is further configured to determine target contour information of the first detection object based on the multiple two-dimensional images;

[0234] The second processing module 1702 is further configured to obtain a range threshold of the multiple two-dimensional images, and perform image segmentation processing on the multiple two-dimensional images according to the range threshold and target contour information to obtain a first contour image.

[0235] The image processing device 1700 in this embodiment extracts contour features from multiple two-dimensional images to obtain target contour information of the first detection object, and then performs image segmentation on the multiple two-dimensional images based on the range threshold and the target contour information to obtain a first contour image of the first detection object, thereby ensuring the information accuracy of the target contour information and thus ensuring the image accuracy of the first contour image.

[0236] In one embodiment of the present application, as shown in FIG21 , an image processing device 2100 is provided. The image processing device 2100 includes a processor 2102 and a memory 2104. The memory 2104 stores a program or instruction that, when executed by the processor 2102, implements the steps of the image processing method described in any of the aforementioned technical solutions. Therefore, the image processing device 2100 has all the beneficial effects of the image processing method described in any of the aforementioned technical solutions, and no further details are given here.

[0237] In one embodiment of the present application, a readable storage medium is provided, on which a program is stored. When the program is executed by a processor, an image processing method as in any of the above embodiments is implemented, thereby having all the beneficial technical effects of the image processing method in any of the above embodiments.

[0238] The readable storage medium includes a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0239] In one embodiment of the present application, a computer program product is provided, comprising computer instructions. When the computer instructions are executed by a processor, the image processing method in any of the above embodiments is implemented, thereby having all the beneficial technical effects of the image processing method in any of the above embodiments.

[0240] In one embodiment of the present application, a computer device is provided, comprising a processor and a memory. The memory stores a program or instructions that, when executed by the processor, implement the steps of the image processing method described in any of the above-described embodiments. Therefore, the computer device possesses all the beneficial effects of the image processing method described in any of the above-described embodiments, and further description thereof is omitted.

[0241] In one embodiment of the present application, an imaging device is proposed, including: an image processing device as in any of the above embodiments, and / or a readable storage medium in any of the above embodiments, and / or a computer program product in any of the above embodiments, and thus has all the beneficial technical effects of the image processing device in any of the above embodiments, and / or the readable storage medium in any of the above embodiments, and / or the computer program product in any of the above embodiments, and no further details will be given here.

[0242] In one embodiment of the present application, the imaging device is any one of the following: a computer tomography device, a magnetic resonance imaging device, or an infrared imaging device.

[0243] In one embodiment of the present application, a surgical robot is proposed, which plans a motion trajectory or avoids obstacles based on a processed image, wherein the image is obtained based on the steps of the image processing method in any of the above embodiments, and thus has all the beneficial technical effects of the image processing method in any of the above embodiments, which will not be elaborated upon here.

[0244] It should be clarified that in the claims, specification and drawings of this application, the term "plurality" refers to two or more. Unless otherwise clearly defined, the orientation or positional relationship indicated by the terms "upper" and "lower" is based on the orientation or positional relationship shown in the drawings. It is only for the purpose of more conveniently describing this application and making the description process simpler, and is not intended to indicate or imply that the device or element referred to must have the specific orientation described, be constructed and operated in a specific orientation. Therefore, these descriptions cannot be understood as limitations on this application. The terms "connect", "install", "fix" and the like should be understood in a broad sense. For example, "connection" can be a fixed connection between multiple objects, or a detachable connection between multiple objects, or an integral connection; it can be a direct connection between multiple objects, or an indirect connection between multiple objects through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood based on the specific circumstances of the above data.

[0245] In the claims, specification, and drawings of this application, the terms "one embodiment," "some embodiments," "a specific embodiment," and the like mean that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of this application. In the claims, specification, and drawings of this application, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0246] The above are merely preferred embodiments of the present application and are not intended to limit the present application. Those skilled in the art will readily appreciate that various modifications and variations are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. An image processing method, wherein: The image processing method comprises: Acquire a three-dimensional image of a first detection object, and convert the three-dimensional image into a plurality of two-dimensional images; Contour extraction is performed on the multiple two-dimensional images to obtain a first contour image of the first detection object.

2. The image processing method according to claim 1, wherein: The step of extracting contours from the plurality of two-dimensional images to obtain a first contour image of the first detection object specifically includes: Determining target contour information of the first detection object through the multiple two-dimensional images; A range threshold of the three-dimensional image is acquired, and image segmentation processing is performed on the multiple two-dimensional images according to the range threshold and the target contour information to obtain the first contour image.

3. The image processing method according to claim 2, wherein: Determining the target contour information of the first detection object through the multiple two-dimensional images specifically includes: Performing contour detection on the multiple two-dimensional images to obtain multiple contour information; Data fusion is performed on the multiple contour information to obtain the target contour information in the multiple contour information.

4. The image processing method according to any one of claims 1 to 3, wherein: The converting the three-dimensional image into a plurality of two-dimensional images specifically comprises: A target pixel axis of the three-dimensional image is acquired, and image conversion processing is performed on the three-dimensional image according to the target pixel axis to obtain the multiple two-dimensional images.

5. The image processing method according to any one of claims 1 to 4, wherein: The obtaining of a three-dimensional image of the first detection object specifically includes: An original image of the first detection object is acquired, and denoising and enhancing processing is performed on the original image to obtain the three-dimensional image.

6. An image processing method, wherein: The image processing method comprises: Acquire a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object according to the three-dimensional image; According to the first contour image, feature extraction is performed on the three-dimensional image to obtain a second contour image of the second detection object.

7. The image processing method according to claim 6, wherein: The step of extracting features from the three-dimensional image according to the first contour image to obtain a second contour image of the second detection object specifically includes: determining non-background pixels in the three-dimensional image according to the first contour image; Acquire a threshold combination, and determine a target threshold in the threshold combination according to the non-background pixel, wherein the threshold combination includes thresholds corresponding to the second detection object of different materials; The three-dimensional image is processed according to the target threshold to determine the second contour image.

8. The image processing method according to claim 7, wherein: Determining the target threshold in the threshold combination according to the non-background pixels specifically includes: Performing data processing on the non-background pixels to obtain a distribution mean of the non-background pixels; The target threshold is determined in the threshold combination by performing a difference operation on the distribution mean and the threshold combination.

9. The image processing method according to claim 7, wherein: The performing image processing on the three-dimensional image according to the target threshold to determine the second contour image specifically includes: Performing image segmentation on the three-dimensional image according to the target threshold to obtain a segmentation result image; A noise filtering process is performed on the segmentation result image to determine the second contour image.

10. The image processing method according to any one of claims 7 to 9, wherein: The determining, according to the first contour image, non-background pixels in the three-dimensional image specifically includes: Determine a pixel set of the first detection object according to the first contour image; Determining background pixels in the three-dimensional image according to the pixel set; According to the background pixels, the pixels in the three-dimensional image are classified to obtain the non-background pixels.

11. The image processing method according to any one of claims 6 to 10, wherein: Determining a first contour image of the first detection object according to the three-dimensional image specifically includes: converting the three-dimensional image into a plurality of two-dimensional images; Contour extraction is performed on the multiple two-dimensional images to obtain the first contour image of the first detection object.

12. The image processing method according to claim 11, wherein: The extracting contours of the multiple two-dimensional images to obtain the first contour image of the first detection object specifically includes: Determining target contour information of the first detection object according to the multiple two-dimensional images; A range threshold of the multiple two-dimensional images is acquired, and image segmentation processing is performed on the multiple two-dimensional images according to the range threshold and the target contour information to obtain the first contour image.

13. An image processing device, wherein: The image processing device comprises: A first processing module, configured to obtain a three-dimensional image of a first detection object and convert the three-dimensional image into a plurality of two-dimensional images; The first processing module is further used to extract contours of the multiple two-dimensional images to obtain a first contour image of the first detection object.

14. An image processing device, wherein: The image processing device comprises: A second processing module is used to obtain a three-dimensional image including a first detection object and a second detection object, where the second detection object is used to carry the first detection object, and determine a first contour image of the first detection object according to the three-dimensional image; The second processing module is further used to extract features from the three-dimensional image according to the first contour image to obtain a second contour image of the second detection object.

15. A readable storage medium, wherein: The readable storage medium stores a program or an instruction, and when the program or the instruction is executed by the processor, the steps of the image processing method according to any one of claims 1 to 5 or the image processing method according to any one of claims 6 to 12 are implemented.

16. A computer program product comprising computer instructions, wherein: When the computer instructions are executed by a processor, the steps of the image processing method according to any one of claims 1 to 5 or the image processing method according to any one of claims 6 to 12 are implemented.

17. A computer device, wherein: include: processor; A memory, wherein a program or instruction is stored in the memory, and the processor implements the steps of the image processing method according to any one of claims 1 to 5 or the image processing method according to any one of claims 6 to 12 when executing the program or instruction in the memory.

18. An imaging device, wherein: The imaging device is any one of the following: a computer tomography device, a magnetic resonance imaging device, an infrared imaging device, and the imaging device includes: The image processing device according to claim 13 or 14; and / or The readable storage medium according to claim 15; and / or A computer program product as claimed in claim 16; and / or The computer device of claim 17.

19. A surgical robot, wherein: The surgical robot plans a motion trajectory or avoids obstacles based on the processed image, wherein the image is an image obtained by processing according to the steps of the image processing method according to any one of claims 1 to 5 or the image processing method according to any one of claims 6 to 12.

Images