WO2022089079A1 - 图像处理方法、装置、设备、系统及计算机可读存储介质 - Google Patents

图像处理方法、装置、设备、系统及计算机可读存储介质 Download PDF

Info

Publication number
WO2022089079A1
WO2022089079A1 PCT/CN2021/118707 CN2021118707W WO2022089079A1 WO 2022089079 A1 WO2022089079 A1 WO 2022089079A1 CN 2021118707 W CN2021118707 W CN 2021118707W WO 2022089079 A1 WO2022089079 A1 WO 2022089079A1
Authority
WO
WIPO (PCT)
Prior art keywords
gray value
total
original
value
pixel
Prior art date
Application number
PCT/CN2021/118707
Other languages
English (en)
French (fr)
Inventor
李炳轩
Original Assignee
湖北锐世数字医学影像科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 湖北锐世数字医学影像科技有限公司 filed Critical 湖北锐世数字医学影像科技有限公司
Publication of WO2022089079A1 publication Critical patent/WO2022089079A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10104—Positron emission tomography [PET]

Definitions

  • the present application relates to the technical field of image processing, and in particular, to an image processing method, apparatus, device, system, and computer-readable storage medium.
  • the quality of medical images such as Positron Emission Computed Tomography (PET) and Computed Tomography (CT) has a decisive influence on the accuracy of clinical diagnosis.
  • PET Positron Emission Computed Tomography
  • CT Computed Tomography
  • noise may be introduced into medical images due to factors such as background radiation, data quality, and reconstruction algorithms.
  • the introduced noise will cause image distortion and lose the quantitative accuracy of the image, and on the other hand, it will drown out the details of the image and reduce the spatial resolution of the image.
  • filtering methods such as mean filtering, median filtering, and Gaussian filtering are usually employed. Although these filtering methods can remove high-frequency noise in most medical images, they will bring about the deterioration of spatial resolution. Therefore, these filtering methods are not ideal noise reduction methods.
  • the purpose of the embodiments of the present application is to provide an image processing method, apparatus, device, system, and computer-readable storage medium, so as to solve at least one problem existing in the prior art.
  • an image processing method which may include:
  • the step of dividing the original image into multiple regions according to a preset rule includes:
  • the original image is divided into regions according to its constituent elements, physical properties or orientation.
  • the step of calculating the total rate of change of the original gray value of each of the regions includes:
  • the total change rate of the original gray value of the region is determined according to the obtained change rate of the original gray value of all the pixel points in the region.
  • the step of determining the total rate of change of the original gray value of the region includes:
  • a mean square error operation is performed on the rate of change of the original grayscale values of all the pixel points in the region to obtain the total rate of change of the original grayscale values of the region.
  • the step of calculating the total rate of change of the original gray value of each of the regions further includes:
  • the step of invoking the constructed first noise reduction model to perform first-order total variation processing on the original gray value includes:
  • first-order gradient processing is performed on the original gray values of all pixel points in all areas where the total change rate of the original gray value is greater than or equal to the preset value:
  • the total gray value is determined as the final gray value of the pixel, Or calculate the final gray value of the pixel point according to the following formula:
  • f′ x,y f x,y +k ⁇ [(f x,y -f x-1,y )+(f x,y -f x,y-1 )],
  • the total gray value is calculated according to the following formula to obtain the final gray value of the pixel point:
  • f x,y , f x-1,y and f x,y-1 represent the pixel points located at coordinates (x,y), (x-1,y) and (x,y-1), respectively.
  • original gray value Represents the total gray value of the pixel located at the coordinate (x, y) obtained after processing
  • f′ x, y represents the final gray value of the pixel located at the coordinate (x, y)
  • x and y are both A natural number
  • k is an image enhancement coefficient.
  • the step of invoking the constructed first noise reduction model to perform first-order total variation processing on the original gray value includes:
  • first-order gradient processing is performed on the original gray values of all pixel points in all areas where the total change rate of the original gray value is greater than or equal to the preset value:
  • the total gray value is determined as the final gray value of the pixel, Or calculate the final gray value of the pixel point according to the following formula:
  • f′ x,y,z f x,y,z +k ⁇ [(f x,y,z -f x-1,y,z )+(f x,y,z -f x,y-1 ,z )+(f x,y,z -f x,y,z-1 )],
  • the total gray value is calculated according to the following formula to obtain the final gray value of the pixel point:
  • f′ x,y,z f x,y,z -(f x,y,z -f x-1,y,z )-(f x,y,z -f x,y-1,z ) -(f x,y,z -f x,y,z-1 ),
  • f x,y,z , f x-1,y,z , f x,y-1,z and f x,y,z-1 are located at coordinates (x,y,z), (x- 1, y, z), (x, y-1, z) and (x, y, z-1) original gray value of the pixel point;
  • f′ x, y, z represents the final gray value of the pixel located at the coordinates (x, y, z) ;
  • x, y and z are all natural numbers;
  • k is the image enhancement coefficient.
  • the step of invoking the second noise reduction model to perform second-order total variational processing on the original gray value includes:
  • second-order gradient processing is sequentially performed on the original gray values of all pixel points in all areas where the total change rate of the original gray value is less than the preset value:
  • the total gray value is determined as the final gray value of the pixel value, or calculate the final gray value of the pixel point according to the following formula:
  • the total gray value is calculated according to the following formula to obtain the final gray value of the pixel point:
  • f x,y , f x-1,y , f x+1,y , f x,y-1 and f x,y+1 are located at coordinates (x,y), (x-1,y respectively ), (x+1, y), (x, y-1) and (x, y+1) original gray value of the pixel point; and Respectively represent the gray value of the pixel at the coordinate (x, y) obtained after processing in the x-axis and y-axis directions; is the total gray value of the pixel located at the coordinate (x, y) obtained after processing; f′ x, y represents the final gray value of the pixel located at the coordinate (x, y); x and y are both A natural number; k is an image enhancement coefficient.
  • the step of invoking the second noise reduction model to perform second-order total variation processing on the original gray value includes:
  • second-order gradient processing is sequentially performed on the original gray values of all pixel points in all areas where the total change rate of the original gray value is less than the preset value:
  • the total gray value is determined as the final gray value of the pixel value, or calculate the final gray value of the pixel point according to the following formula:
  • the total gray value is calculated according to the following formula to obtain the final gray value of the pixel point:
  • f x,y,z , f x-1,y,z , f x+1,y-1,z , f x,y-1,z , f x,y+1,z , f x, y, z-1 and f x, y, z+1 represent coordinates (x, y, z), (x-1, y, z), (x+1, y, z), (x, y), respectively -1, z), (x, y+1, z) (x, y, z-1), (x, y, z+1)
  • the original grayscale value of the pixel point and respectively represent the grayscale values of the pixels located at the coordinates (x, y, z) in the x-axis, y-axis and z-axis directions obtained after processing; is the total gray value of the pixel at coordinates (x, y, z) obtained after processing; f′ x, y, z represents the final
  • the original images include CT images, MRI images, PET images, PET-CT images or ultrasound images.
  • the embodiment of the present application also provides an image processing apparatus, and the apparatus may include:
  • a dividing unit which is configured to divide the acquired original image into a plurality of regions according to preset rules
  • a calculation unit configured to calculate the total rate of change of the original gray value of each of said regions
  • a processing unit configured to judge whether the total change rate is greater than a preset value, and when judging that the total change rate of the original gray value is greater than or equal to the preset value, call the constructed first noise reduction
  • the model performs first-order total variation processing on the original gray value, and when it is judged that the total change rate of the original gray value is less than the preset value, the constructed second noise reduction model is called for the The original gray value is subjected to second-order total variation processing.
  • Embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, the above-mentioned image processing method can be implemented.
  • An embodiment of the present application further provides a computer device, which may include a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed, the processor executes the above image processing method .
  • Embodiments of the present application further provide an image processing system, which may include the above computer device and a scanning device connected to the above computer device.
  • the scanning device includes a CT scanner, an MRI scanner, a PET detector, a PET-CT device or an ultrasound device.
  • the embodiments of the present application divide the original image into multiple areas, calculate the total change rate of the original gray value of each area, and calculate the total change rate of the original gray value of each area according to the original gray value of each area. According to the relationship between the total change rate and the preset value, select the first noise reduction model to perform first-order total variation processing on the original gray value or select the second noise reduction model to perform second-order total variation processing on the original gray value, Therefore, the purpose of improving the spatial resolution of the image can be achieved while removing the noise in the image.
  • FIG 1 is an application environment diagram of an image processing method in an embodiment of the present application
  • FIG. 2 is a schematic flowchart of an image processing method provided by an embodiment of the present application.
  • FIG. 3 is a schematic structural diagram of an image processing apparatus provided by an embodiment of the present application.
  • FIG. 4 is a schematic structural diagram of a computer device provided by an embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of a computer device provided by another embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of an image processing system provided by an embodiment of the present application.
  • the image processing method provided by the embodiment of the present application may be applicable to the application environment shown in FIG. 1 .
  • the method can be applied to computer equipment.
  • the computer equipment includes a terminal 1000 and a server 2000 connected through a network.
  • the method can be executed in the terminal 1000 or the server 2000.
  • the terminal 1000 can directly obtain the original image of the target object (for example, a person or pet, etc.) from the scanning device or the photographing device, and execute the image processing method on the terminal side; or,
  • the terminal 1000 may also send the original image of the target object to the server 2000 after obtaining the original image, so that the server 2000 can obtain the original image of the target object and execute the image processing method.
  • the terminal 1000 may specifically be a desktop terminal (eg, a desktop computer) or a mobile terminal (eg, a notebook computer, a tablet computer, a mobile phone, or a personal digital assistant).
  • the server 2000 can be implemented by an independent server or a server cluster composed of multiple servers.
  • the execution body may be an image processing apparatus, and the image processing apparatus may be implemented as a part or all of the above-mentioned terminal or server through software, hardware, or a combination of software and hardware.
  • FIG. 2 is an image processing method provided by an embodiment of the present application, the method can be executed by an image processing apparatus, and can include the following steps:
  • S1 Divide the acquired original image into multiple regions according to preset rules.
  • the original image may be a medical image such as a CT image, a PET image, a PET-CT image, a Magnetic Resonance Imaging (MRI) image or an ultrasound image, or may be any other type of image.
  • a medical image such as a CT image, a PET image, a PET-CT image, a Magnetic Resonance Imaging (MRI) image or an ultrasound image, or may be any other type of image.
  • MRI Magnetic Resonance Imaging
  • the original image may be divided into multiple regions according to preset rules.
  • the original image can be divided into multiple areas according to the constituent elements of the original image.
  • the original image can be divided into head area, upper body area and lower body area, etc.; it can also be divided according to the brightness, contrast and / or physical properties such as saturation (or gray value) to divide it into multiple areas; it is also possible to divide the original image into two areas, such as upper area and lower area, or upper area, lower area, left area, Four areas such as the right area, but not limited to this.
  • the total change rate is mainly used to indicate the overall change rule of the original grayscale values of all pixel points in the area.
  • the preset value can be set according to empirical data or application scenarios, and the value is generally small.
  • the total change rate of the original gray value of each region can be calculated in turn and judged whether it is greater than the preset value. Specifically, for each area, the difference between the original gray values of each two adjacent pixels can be calculated in turn, and the difference between the original gray values of each two adjacent pixels and its coordinates can be obtained. The difference value is used to determine the rate of change of the original gray value of the two pixel points, and the total rate of change of the original gray value of the region is determined according to the rate of change of the original gray value of all the obtained pixel points.
  • the average value of the rate of change of the original gray values of all pixels may be determined as the total rate of change of the original gray values of the region, or the average rate of change of the original gray values of all pixels may also be averaged.
  • the variance operation is used to obtain the total change rate of the original gray value of the region, but it is not limited to the above method.
  • the total change rate of the original gray value of the area can also be determined by counting the change rule of the original gray value of all pixel points. For example, when the original grayscale values of all pixel points change linearly (preferably, smoothly linearly change), the slope of the linear change can be determined as the total change rate. At this time, it can also be understood that the change rates of the original grayscale values of all pixel points are equal, and the total change rate is the change rate.
  • S3 Determine whether the total rate of change of the original gray value of each area is greater than the preset value, and when it is determined that the rate of change of the original gray value is greater than the preset value, call the constructed first noise reduction model for the original gray value First-order total variation processing is performed on the original gray value, and when it is determined that the total change rate of the original gray value is less than the preset value, the constructed second noise reduction model is called to perform second-order total variation processing on the original gray value.
  • the first noise reduction model and the second noise reduction model may respectively refer to models capable of performing first-order total variation processing and second-order total variation processing on various image data (including medical images).
  • the constructed first noise reduction model After calculating the total change rate of the original gray value of each area, it can be judged in turn whether the total change rate of the original gray value of each area is greater than the preset value, when it is judged that the change rate of the original gray value is greater than
  • the constructed first noise reduction model can be called to perform first-order total variation processing on the original gray value, and when it is determined that the total change rate of the original gray value is less than the preset value, the constructed first noise reduction model can be called.
  • the second noise reduction model performs second-order total variation processing on the original gray value.
  • both the first preset threshold and the second preset threshold can be set according to the original gray value, and the first preset threshold can be greater than 0 and less than satisfying the condition that the total change rate of the original gray value is greater than or equal to the preset value
  • the second preset threshold value may be greater than 0 and less than the maximum value of the original grayscale values in all regions satisfying the condition that the total change rate of the grayscale value is less than the preset value.
  • the step of invoking the constructed first noise reduction model to perform first-order total variation processing on the original gray value can be performed according to the following process:
  • the original gray value of all pixels in all regions that satisfy the condition that the total change rate of the original gray value is greater than or equal to the preset value can be sequentially performed according to the following formula (1).
  • Step gradient processing for each area that satisfies the conditions, determine whether the total gray value of each pixel obtained after processing is greater than the first preset threshold; for each pixel, when the total gray value of the pixel is determined When it is greater than or equal to the first preset threshold, the total gray value obtained after processing is determined as the final gray value of the pixel, or the final gray value of the pixel is calculated according to the following formula (2), and when When it is determined that the total gray value obtained after processing is smaller than the first preset threshold, the total gray value obtained after processing is calculated according to the following formula (3) to obtain the final gray value of the pixel.
  • f x,y , f x-1,y and f x,y-1 represent the pixel points located at coordinates (x,y), (x-1,y) and (x,y-1), respectively.
  • original gray value Represents the total gray value of the pixel located at the coordinate (x, y) obtained after the first-order gradient processing
  • f′ x, y represents the final gray value of the pixel located at the coordinate (x, y)
  • x and y are natural numbers
  • k is the image enhancement coefficient, and its value is generally within 10-5 to 105 .
  • the original gray value of all pixels in all regions that satisfy the condition that the total rate of change of the original gray value is greater than or equal to the preset value can be performed step by step according to the following formula (4). For each area that satisfies the conditions, determine whether the total gray value of each pixel obtained after processing is greater than the first preset threshold; for each pixel, when it is determined that the total gray value obtained after processing is When the degree value is greater than or equal to the first preset threshold, the total gray value obtained after processing is determined as the final gray value of the pixel, or the final gray value of the pixel is calculated according to the following formula (5), When it is determined that the total gray value obtained after processing is smaller than the first preset threshold, the total gray value obtained after processing is calculated according to the following formula (6) to obtain the final gray value of the pixel.
  • f′ x,y,z f x,y,z -(f x,y,z -f x-1,y,z )-(f x,y,z -f x,y-1,z ) -(f x,y,z -f x,y,z-1 ) (6)
  • f x,y,z , f x-1,y,z , f x,y-1,z and f x,y,z-1 are located at coordinates (x,y,z), (x- 1, y, z), (x, y-1, z) and (x, y, z-1) original gray value of the pixel point;
  • f′ x, y, z represents the pixel at the coordinate (x, y, z)
  • the total gray value of the point Final grayscale value; x, y, and z are all natural numbers.
  • the noise in the region can be removed, and compared with the medium-pass in the prior art Filtering, mean filtering and Gaussian filtering can improve its spatial resolution.
  • the contrast and resolution of the image can be improved.
  • the step of invoking the constructed second noise reduction model to perform second-order total variation processing on the original gray value can be performed according to the following process:
  • a second-order gradient can be performed on the original grayscale values of all pixels in all regions that satisfy the condition that the total change rate of the original grayscale value is less than the preset value according to the following formula (7). processing; for each area that satisfies the conditions, determine whether the total gray value of each pixel obtained after processing is greater than the second preset threshold; for each pixel, when it is determined that the total gray value obtained after processing is When the value is greater than or equal to the second preset threshold, the total gray value obtained after processing is determined as the final gray value of the pixel, or the final gray value of the pixel is calculated according to the following formula (8), and When it is determined that the total gray value obtained after processing is smaller than the second preset threshold, the total gray value obtained after processing is calculated according to the following formula (9) to obtain the final gray value of the pixel.
  • f x,y , f x-1,y , f x+1,y , f x,y-1 and f x,y+1 are located at coordinates (x,y), (x-1,y respectively ), (x+1, y), (x, y-1) and (x, y+1) original gray value of the pixel point; and respectively represent the gray value of the pixel at the coordinates (x, y) in the x-axis and y-axis directions obtained after the second-order gradient processing; is the total gray value of the pixel located at the coordinate (x, y) obtained after the second-order gradient processing; f′ x, y represents the final gray value of the pixel located at the coordinate (x, y).
  • second-order gradient processing can be performed on the original grayscale values of all pixels in all regions that satisfy the condition that the total change rate of the original grayscale value is less than the preset value according to the following formula (10). For each area that satisfies the condition, judge whether the total gray value of each pixel obtained after processing is greater than the second preset threshold; For each pixel, when it is judged that its total gray value is greater than or equal to When the second preset threshold is used, the total gray value is determined as the final gray value of the pixel, or the final gray value of the pixel is calculated according to the following formula (11), and when the total gray value is determined When the value is less than the second preset threshold, the final gray value of the pixel is obtained by calculating the total gray value obtained after processing according to the following formula (12).
  • f x,y,z , f x-1,y,z , f x+1,y-1,z , f x,y-1,z , f x,y+1,z , f x, y, z-1 and f x, y, z+1 represent coordinates (x, y, z), (x-1, y, z), (x+1, y, z), (x, y), respectively -1, z), (x, y+1, z) (x, y, z-1), (x, y, z+1)
  • the original grayscale value of the pixel point and respectively represent the gray value of the pixel at the coordinates (x, y, z) in the x-axis, y-axis and z-axis directions obtained after the second-order gradient processing; is the total gray value of the pixel at coordinates (x, y, z) obtained after second-order gradient processing; f′ x,
  • the spatial resolution can be improved, and the Guarantee the authenticity of the image.
  • the gray value of the image changes smoothly and linearly and the slope of its linear change is small
  • the second noise reduction model to perform second-order total variation processing on the original gray value
  • the spatial resolution can be improved, and the Guarantee the authenticity of the image.
  • the gray value of the pixel in all directions is 0, so that the final gray value of the pixel is equal to its original gray value.
  • the contrast and resolution of the image can be improved.
  • the coordinates (x-1,y), (x+1,y), (x,y-1) and (x,y+1) in the above formula only represent the ), which may not refer to specific coordinate values.
  • the pixels at coordinates (x-1,y) and (x+1,y) are adjacent to the pixel at coordinates (x,y) in the x-axis direction
  • the pixels at coordinates (x,y- 1) are adjacent to the pixel at coordinates (x, y) in the y-axis direction.
  • the coordinates (x-1,y,z), (x+1,y,z), (x,y-1,z), (x,y+1,z)(x,y) in the above formula , z-1), (x, y, z+1) also only represent the positional relationship with the coordinates (x, y, z).
  • the embodiment of the present application divides the original image into multiple regions, calculates the total change rate of the original gray value of each region, and calculates the total change rate of the original gray value of each region according to the difference between the total change rate of the original gray value of each region and the
  • the magnitude relationship of the preset value is to select the first noise reduction model to perform first-order total variation processing on the original gray value or select the second noise reduction model to perform second-order total variation processing on the original gray value.
  • the first-order gradient processing result is compared with the first preset threshold, or the second-order gradient result is compared with the second preset threshold, which makes it possible to remove the grayscale value lower than the first preset threshold or the second preset threshold.
  • the noise at the position where the threshold is set, and the position higher than the first preset threshold or the second preset threshold will continue to maintain and enhance the original gradient, so that the gradient of the position with high gray value is higher, and the gradient of the position with low gray value will be higher.
  • the local gradient is lower, which enhances the contrast and improves the resolution, so that the purpose of improving the spatial resolution of the image can be achieved while removing the noise in the image.
  • the embodiment of the present application also provides an image processing apparatus, as shown in FIG. 3 , which may include:
  • a dividing unit 310 which is configured to divide the acquired original image into a plurality of regions according to preset rules
  • a calculation unit 320 which is configured to calculate the total rate of change of the original gray value of each region
  • the processing unit 330 is configured to determine whether the total rate of change is greater than a preset value, and when determining that the total rate of change of the original grayscale value is greater than or equal to the preset value, call the constructed first noise reduction model for the original grayscale value. First-order total variation processing is performed on the original gray value, and when it is determined that the total change rate of the original gray value is less than the preset value, the constructed second noise reduction model is called to perform second-order total variation processing on the original gray value.
  • FIG. 4 shows a schematic structural diagram of a computer device in an embodiment.
  • the computer device may be the terminal 1000 in FIG. 1 .
  • the computer device includes a processor, memory, network interface, input device and display connected through a system bus.
  • the memory includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium of the computer device may store an operating system, and may also store a computer program.
  • the processor may execute the image processing method described in the above embodiments.
  • a computer program may also be stored in the internal memory.
  • the image processing method described in the above embodiments is executed.
  • FIG. 5 shows a schematic structural diagram of a computer device in another embodiment.
  • the computer device may specifically be the server 2000 in FIG. 1 .
  • the computer device includes a processor, memory, and a network interface connected through a system bus.
  • the memory includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium of the computer device stores an operating system, and may also store a computer program.
  • the processor can execute the image processing method described in the above embodiments.
  • a computer program may also be stored in the internal memory.
  • the image processing method described in the above embodiments is executed.
  • FIG. 4 and FIG. 5 are only block diagrams of partial structures related to the solution of the present application, and do not constitute a limitation on the computer equipment to which the solution of the present application is applied.
  • a computer device may include more or fewer components than those shown in the figures, or combine certain components, or have a different configuration of components.
  • the present application further provides an image processing system
  • the image processing system may include the computer device shown in FIG. 4 or FIG. 5 and a scanning device connected thereto, and the scanning device may use It is used to obtain an original image by scanning a target object and provide the obtained original image to a computer device.
  • the scanning device may be any device capable of detecting radioactive rays, for example, it may include a CT scanner, an MRI scanner, a PET detector, a PET-CT device or an ultrasound device, etc., but not limited thereto.
  • the present application further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed, the corresponding functions described in the foregoing method embodiments can be implemented.
  • the computer program can also run on a computer device as shown in FIG. 4 or FIG. 5 .
  • the memory of the computer device contains each program module that constitutes the device, and the computer program formed by each program module can realize the functions corresponding to each step in the image segmentation method described in the above embodiments when executed.
  • Nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Road (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
  • SRAM static RAM
  • DRAM dynamic RAM
  • SDRAM synchronous DRAM
  • DDRSDRAM double data rate SDRAM
  • ESDRAM enhanced SDRAM
  • SLDRAM synchronous chain Road (Synchlink) DRAM
  • SLDRAM synchronous chain Road (Synchlink) DRAM
  • Rambus direct RAM
  • DRAM direct memory bus dynamic RAM
  • RDRAM memory bus dynamic RAM
  • the systems, devices, devices, units, etc. described in the above embodiments may be specifically implemented by semiconductor chips, computer chips and/or entities, or by products with certain functions.
  • the functions are divided into various units and described respectively.
  • the functions of each unit may be implemented in one or more chips.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

一种图像处理方法、装置、设备、系统及计算机可读存储介质，该方法包括：按照预设规则将所获取的原始图像划分为多个区域（S1）；计算每个所述区域的原始灰度值的总变化率（S2）；判断所述总变化率是否大于预设值，当判断出所述原始灰度值的总变化率大于或等于所述预设值时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理，并且当判断出所述原始灰度值的总变化率小于所述预设值时，调用所构建的第二降噪模型对所述原始灰度值进行二阶全变分处理（S3）。通过上述方法，可以实现在去除图像中噪声的同时提高其空间分辨率的目的。

Description

图像处理方法、装置、设备、系统及计算机可读存储介质
本申请要求于2020年10月30日提交的中国专利申请202011193184.X的优先权，其全部内容通过援引加入本文。
技术领域
本申请涉及图像处理技术领域，特别涉及一种图像处理方法、装置、设备、系统及计算机可读存储介质。
背景技术
正电子发射断层成像(Positron Emission Computed Tomography，简称PET)、计算机断层成像(Computed Tomography，简称CT)等医学图像的质量对临床诊断的准确度具有决定性的影响。然而，在医学成像过程中，由于背景辐射、数据质量、重建算法等因素的影响，可能会为医学图像引入噪声。所引入的噪声一方面会导致图像失真，损失图像的定量精度，另一方面还会淹没图像的细节信息，降低图像的空间分辨率。
为了去除像PET图像或CT图像等医学图像中的噪声，通常采用均值滤波、中值滤波、高斯滤波等滤波方法。这些滤波方法虽然可以消除大部分医学图像中的高频噪声，但是会带来空间分辨率的恶化，因此，这些滤波方法并不是一种理想的降噪方法。
因此，需要一种新的图像处理方法来去除医学图像或其它类型的图像中的噪声，并且提高图像的空间分辨率。
发明内容
本申请实施例的目的在于提供一种图像处理方法、装置、设备、系统及计 算机可读存储介质，以解决现有技术中存在的至少一种问题。
为解决上述技术问题，本申请实施例提供了一种图像处理方法，该方法可以包括：
按照预设规则将所获取的原始图像划分为多个区域；
计算每个所述区域的原始灰度值的总变化率；
判断所述总变化率是否大于预设值，当判断出所述原始灰度值的总变化率大于或等于所述预设值时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理，并且当判断出所述原始灰度值的总变化率小于所述预设值时，调用所构建的第二降噪模型对所述原始灰度值进行二阶全变分处理。
可选地，按照预设规则将所述原始图像划分为多个区域的步骤包括：
根据所述原始图像的构成元素、物理属性或方位将所述原始图像划分为多个区域。
可选地，计算每个所述区域的原始灰度值的总变化率的步骤包括：
针对每个所述区域，依次计算所述区域内的每两个相邻像素点的原始灰度值之间的差值；
根据所述原始灰度值的差值与每两个相邻所述像素点的坐标差值来确定两个相邻所述像素点的所述原始灰度值的变化率；
根据所得到的所述区域内的所有所述像素点的所述原始灰度值的变化率来确定所述区域的所述原始灰度值的总变化率。
可选地，确定所述区域的所述原始灰度值的总变化率的步骤包括：
将所述区域内的所有所述像素点的所述原始灰度值的变化率的平均值确定为所述区域的所述原始灰度值的总变化率；或者
对所述区域内的所有所述像素点的所述原始灰度值的变化率进行均方差运算以获得所述区域的所述原始灰度值的总变化率。
可选地，计算每个所述区域的原始灰度值的总变化率的步骤还包括：
统计每个所述区域内的所有像素点的所述原始灰度值的变化规律，并且当所述原始灰度值呈线性变化时，将所述原始灰度值的线性变化的斜率确定为所述区域的所述原始灰度值的所述总变化率。
可选地，当所述原始图像为二维图像时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理的步骤包括：
按照以下公式依次对满足所述原始灰度值的总变化率大于或等于所述预设值的所有区域内的所有像素点的原始灰度值进行一阶梯度处理：
Figure PCTCN2021118707-appb-000001
针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大于第一预设阈值；
针对每个像素点，当判断出处理后得到的所述总灰度值大于或等于所述第一预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按照以下公式来计算所述像素点的最终灰度值：
f′ x,y＝f x,y+k×[(f x,y-f x-1,y)+(f x,y-f x,y-1)]，
而当判断出所述总灰度值小于所述第一预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
f′ x,y＝f x,y-(f x,y-f x-1,y-(f x,y-f x,y-1)，
其中，f x,y、f x-1,y和f x,y-1分别表示位于坐标(x,y)、(x-1,y)和(x,y-1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000002
表示处理后得到的位于坐标(x,y)处的像素点的总灰度值；f′ x,y表示位于坐标(x,y)处的像素点的最终灰度值；x和y均为自然数；k为图像增强系数。
可选地，当所述原始图像为三维图像时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理的步骤包括：
按照以下公式依次对满足所述原始灰度值的总变化率大于或等于所述预设值的所有区域内的所有像素点的原始灰度值进行一阶梯度处理：
Figure PCTCN2021118707-appb-000003
针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大于第一预设阈值；
针对每个像素点，当判断出处理后得到的所述总灰度值大于或等于所述第一预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按照以下公式来计算所述像素点的最终灰度值：
f′ x,y,z＝f x,y,z+k×[(f x,y,z-f x-1,y,z)+(f x,y,z-f x,y-1,z)+(f x,y,z-f x,y,z-1)]，
而当判断出所述总灰度值小于所述第一预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
f′ x,y,z＝f x,y,z-(f x,y,z-f x-1,y,z)-(f x,y,z-f x,y-1,z)-(f x,y,z-f x,y,z-1)，
其中，f x,y,z、f x-1,y,z、f x,y-1,z和f x,y,z-1分别表示位于坐标(x,y,z)、(x-1,y,z)、(x,y-1,z)和(x,y,z-1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000004
表示处理后得到的位于坐标(x,y,z)处的像素点的总灰度值；f′ x,y,z表示位于坐标(x,y,z)处的像素点的最终灰度值；x、y和z均为自然数；k为图像增强系数。
可选地，当所述原始图像为二维图像时，调用所述第二降噪模型对所述原始灰度值进行二阶全变分处理的步骤包括：
按照以下公式依次对满足所述原始灰度值的总变化率小于所述预设值的所有区域内的所有像素点的原始灰度值进行二阶梯度处理：
Figure PCTCN2021118707-appb-000005
Figure PCTCN2021118707-appb-000006
Figure PCTCN2021118707-appb-000007
针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大于第二预设阈值；
针对每个所述像素点，当判断出处理后得到的所述总灰度值大于或等于所述第二预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按 照以下公式来计算所述像素点的最终灰度值：
Figure PCTCN2021118707-appb-000008
而当判断出所述总灰度值小于所述第二预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
Figure PCTCN2021118707-appb-000009
其中，f x,y、f x-1,y、f x+1,y、f x,y-1和f x,y+1分别表示位于坐标(x,y)、(x-1,y)、(x+1,y)、(x,y-1)和(x,y+1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000010
和
Figure PCTCN2021118707-appb-000011
分别表示处理后得到的位于坐标(x,y)处的像素点在x轴和y轴方向上的灰度值；
Figure PCTCN2021118707-appb-000012
为处理后得到的位于坐标(x,y)处的像素点的总灰度值；f′ x,y表示位于坐标(x,y)处的像素点的最终灰度值；x和y均为自然数；k为图像增强系数。
可选地，当所述原始图像为三维图像时，调用所述第二降噪模型对所述原始灰度值进行二阶全变分处理的步骤包括：
按照以下公式依次对满足所述原始灰度值的总变化率小于所述预设值的所有区域内的所有像素点的原始灰度值进行二阶梯度处理：
Figure PCTCN2021118707-appb-000013
Figure PCTCN2021118707-appb-000014
Figure PCTCN2021118707-appb-000015
Figure PCTCN2021118707-appb-000016
针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大于第二预设阈值；
针对每个所述像素点，当判断出处理后得到的所述总灰度值大于或等于所述第二预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按照以下公式来计算所述像素点的最终灰度值：
Figure PCTCN2021118707-appb-000017
而当判断出所述总灰度值小于所述第二预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
Figure PCTCN2021118707-appb-000018
其中，f x,y,z、f x-1,y,z、f x+1,y-1,z、f x,y-1,z、f x,y+1,z、f x,y,z-1和f x,y,z+1分别表示位于坐标(x,y,z)、(x-1,y,z)、(x+1,y,z)、(x,y-1,z)、(x,y+1,z)(x,y,z-1)、(x,y,z+1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000019
和
Figure PCTCN2021118707-appb-000020
分别表示处理后得到的位于坐标(x,y,z)处的像素点在x轴、y轴和z轴方向上的灰度值；
Figure PCTCN2021118707-appb-000021
为处理后得到的位于坐标(x,y,z)处的像素点的总灰度值；f′ x,y,z表示位于坐标(x,y,z)处的像素点的最终灰度值；x、y和z均为自然数；k为图像增强系数。
可选地，所述原始图像包括CT图像、MRI图像、PET图像、PET-CT图像或超声图像。
本申请实施例还提供了一种图像处理装置，该装置可以包括：
划分单元，其被配置为按照预设规则将所获取的原始图像划分为多个区域；
计算单元，其被配置为计算每个所述区域的原始灰度值的总变化率；
处理单元，其被配置为判断所述总变化率是否大于预设值，在判断出所述原始灰度值的总变化率大于或等于所述预设值时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理，并且在判断出所述原始灰度值的总变化率小于所述预设值时，调用所构建的第二降噪模型对所述原始灰度值进行二阶全变分处理。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被执行时能够实现上述图像处理方法。
本申请实施例还提供了一种计算机设备，该设备可以包括处理器和存储器，其中，所述存储器上存储有计算机程序，当所述计算机程序被执行时，所述处理器执行上述图像处理方法。
本申请实施例还提供了一种图像处理系统，该系统可以包括上述计算机设备以及与上述计算机设备连接的扫描设备。
可选地，所述扫描设备包括CT扫描仪、MRI扫描仪、PET探测器、PET- CT设备或超声设备。
由以上本申请实施例提供的技术方案可见，本申请实施例通过将原始图像划分为多个区域，计算每个区域的原始灰度值的总变化率，并且根据每个区域的原始灰度值的总变化率与预设值的大小关系来选择第一降噪模型对原始灰度值进行一阶全变分处理或选择第二降噪模型对原始灰度值进行二阶全变分处理，从而可以实现在去除图像中的噪声的同时提高其空间分辨率的目的。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请中记载的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是本申请的一个实施例中的图像处理方法的应用环境图；
图2是本申请的一个实施例提供的图像处理方法的流程示意图；
图3是本申请的一个实施例提供的图像处理装置的结构示意图；
图4是本申请的一个实施例提供的计算机设备的结构示意图；
图5是本申请的另一个实施例提供的计算机设备的结构示意图；
图6是本申请的一个实施例提供的图像处理系统的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是用于解释说明本申请的一部分实施例，而不是全部的实施例，并不希望限制本申请的范围或权利要求书。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动的前提下所获得的所有其它实施例，都应当属于本申请保护的范围。
需要说明的是，当元件被称为“设置在”另一个元件上，它可以直接设置在另一个元件上或者也可以存在居中的元件。当元件被称为“连接/联接”至另一个元件，它可以是直接连接/联接至另一个元件或者可能同时存在居中元件。本文所使用的术语“连接/联接”可以包括电气和/或机械物理连接/联接。本文所使用的术语“包括/包含”指特征、步骤或元件的存在，但并不排除一个或更多个其它特征、步骤或元件的存在或添加。本文所使用的术语“和/或”包括一个或多个相关所列项目的任意的和所有的组合。
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同。本文中所使用的术语只是为了描述具体实施例的目的，而并不是旨在限制本申请。
另外，在本申请的描述中，术语“第一”、“第二”、“第三”等仅用于描述目的和区别类似的对象，两者之间并不存在先后顺序，也不能理解为指示或暗示相对重要性。此外，在本申请的描述中，除非另有说明，“多个”的含义是两个或两个以上。
本申请实施例提供的图像处理方法可以适用于如图1所示的应用环境中。该方法可以应用于计算机设备。该计算机设备包括通过网络连接的终端1000和服务器2000。该方法可以在终端1000或服务器2000中执行，例如，终端1000可直接从扫描设备或拍摄设备获取目标对象(例如，人或宠物等)的原始图像，并在终端侧执行图像处理方法；或者，终端1000也可在获取目标对象的原始图像后将该原始图像发送至服务器2000，使得服务器2000能够获得目标对象的原始图像并执行图像处理方法。终端1000具体可以是台式终端(例如，台式电脑)或移动终端(例如，笔记本电脑、平板电脑、手机或个人数字助理)。服务器2000可以用独立的服务器或者是多个服务器组成的服务器集群来实现。
需要说明的是，本申请实施例提供的图像处理方法，其执行主体可以是图像处理装置，该图像处理装置可以通过软件、硬件或者软硬件结合的方式实现 为上述终端或服务器的一部分或全部。
下面结合具体实施例对本申请提供的图像处理方法进行详细说明。
图2为本申请的一个实施例提供的图像处理方法，该方法可以由图像处理装置执行，并且可以包括如下步骤：
S1：按照预设规则将所获取的原始图像划分为多个区域。
原始图像可以是CT图像、PET图像、PET-CT图像、磁共振成像(Magnetic Resonance Imaging，简称MRI)图像或超声图像等医学图像，也可以是其它任意类型的图像。
在获取到待处理的原始图像之后，可以按照预设规则将原始图像划分为多个区域。例如，可以按照原始图像的构成元素将原始图像划分为多个区域，例如，对于人体图像，可以将其划分为头部区域、上身区域和下身区域等；还可以按照原始图像的亮度、对比度和/或饱和度(或灰度值)等物理属性将其划分为多个区域；也还可以按照方位将原始图像划分为上区域和下区域等两个区域或者上区域、下区域、左区域、右区域等四个区域，但不限于此。
S2：计算每个区域的原始灰度值的总变化率。
所述总变化率主要用于指示区域内的所有像素点的原始灰度值的整体变化规律。所述预设值可以根据经验数据或应用场景来设定，一般数值较小。
在将原始图像划分为多个区域之后，可以依次计算每个区域的原始灰度值的总变化率并判断其是否大于预设值。具体地，针对每个区域，可以依次计算每两个相邻像素点的原始灰度值之间的差值，根据所得到的每两个相邻像素点的原始灰度值的差值与其坐标差值来确定这两个像素点的原始灰度值的变化率，并且根据所有得到的所有像素点的原始灰度值的变化率来确定该区域的原始灰度值的总变化率。例如，可以将所有像素点的原始灰度值的变化率的平均值确定为该区域的原始灰度值的总变化率，或者也还可以对所有像素点的原始灰度值的变化率进行均方差运算来获得该区域的原始灰度值的总变化率，但不限于 上述方式。另外，针对每个区域，也可以通过统计所有像素点的原始灰度值的变化规律来确定该区域的原始灰度值的总变化率。例如，当所有像素点的原始灰度值呈线性变化(优选地，平滑线性变化)，则可以将其线性变化的斜率确定为总变化率。此时，也可以理解为所有像素点的原始灰度值的变化率相等，该总变化率即为变化率。
S3：判断每个区域的原始灰度值的总变化率是否大于预设值，当判断出原始灰度值的变化率大于预设值时，调用所构建的第一降噪模型对原始灰度值进行一阶全变分处理，并且当判断出原始灰度值的总变化率小于预设值时，调用所构建的第二降噪模型对原始灰度值进行二阶全变分处理。
第一降噪模型和第二降噪模型可以分别是指能够对各种图像数据(包括医学图像)进行一阶全变分处理和二阶全变分处理的模型。
在计算出每个区域的原始灰度值的总变化率之后，可以依次判断每个区域的原始灰度值的总变化率是否大于预设值，当判断出其原始灰度值的变化率大于预设值时，可以调用所构建的第一降噪模型对原始灰度值进行一阶全变分处理，并且当判断出原始灰度值的总变化率小于预设值时，调用所构建的第二降噪模型对原始灰度值进行二阶全变分处理。其中，第一预设阈值和第二预设阈值均可以根据原始灰度值来设置，第一预设阈值可以大于0并且小于满足原始灰度值的总变化率大于或等于预设值这个条件的所有区域内的原始灰度值的最大值，第二预设阈值可以大于0并且小于满足灰度值的总变化率小于预设值这个条件的所有区域内的原始灰度值的最大值。
可以按照如下过程来执行所述调用所构建的第一降噪模型对原始灰度值进行一阶全变分处理的步骤：
当原始图像为二维图像时，可以按照以下公式(1)依次对满足原始灰度值的总变化率大于或等于预设值这个条件的所有区域内的所有像素点的原始灰度值进行一阶梯度处理；针对满足所述条件的每个区域，判断处理后得到的每 个像素点的总灰度值是否大于第一预设阈值；针对每个像素点，当判断出其总灰度值大于或等于第一预设阈值时，将处理后得到的总灰度值确定为该像素点的最终灰度值，或者按照以下公式(2)来计算该像素点的最终灰度值，而当判断出处理后得到的总灰度值小于第一预设阈值时，按照以下公式(3)对处理后得到的总灰度值进行计算来获得该像素点的最终灰度值。
Figure PCTCN2021118707-appb-000022
f′ x,y＝f x,y+k×[(f x,y-f x-1,y)+(f x,y-f x,y-1)]       (2)
f′ x,y＝f x,y-(f x,y-f x-1,y)-(f x,y-f x,y-1)           (3)
其中，f x,y、f x-1,y和f x,y-1分别表示位于坐标(x,y)、(x-1,y)和(x,y-1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000023
表示经过一阶梯度处理后得到的位于坐标(x,y)处的像素点的总灰度值；f′ x,y表示位于坐标(x,y)处的像素点的最终灰度值；x和y均为自然数；k为图像增强系数，其值一般在10- 5～10 5内。
当原始图像为三维图像时，可以按照以下公式(4)依次对满足原始灰度值的总变化率大于或等于预设值这个条件的所有区域内的所有像素点的原始灰度值进行一阶梯度处理；针对满足所述条件的每个区域，判断处理后得到的每个像素点的总灰度值是否大于第一预设阈值；针对每个像素点，当判断出处理后得到的总灰度值大于或等于第一预设阈值时，将处理后得到的总灰度值确定为该像素点的最终灰度值，或者按照以下公式(5)来计算该像素点的最终灰度值，而当判断出处理后得到的总灰度值小于第一预设阈值时，按照以下公式(6)对处理后得到的总灰度值进行计算来获得该像素点的最终灰度值。
Figure PCTCN2021118707-appb-000024
f′ x,y,z＝f x,y,z+k×[(f x,y,z-f x-1,y,z)+(f x,y,z-f x,y-1,z)+(f x,y,z-f x,y,z-1)]   (5)
f′ x,y,z＝f x,y,z-(f x,y,z-f x-1,y,z)-(f x,y,z-f x,y-1,z)-(f x,y,z-f x,y,z-1)      (6)
其中，f x,y,z、f x-1,y,z、f x,y-1,z和f x,y,z-1分别表示位于坐标(x,y,z)、(x-1,y,z)、(x,y-1,z)和(x,y,z-1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000025
表示经过一阶 梯度处理后得到的位于坐标(x,y,z)处的像素点的总灰度值；f′ x,y,z表示位于坐标(x,y,z)处的像素点的最终灰度值；x、y和z均为自然数。
通过调用所构建的第一降噪模型对灰度值变化较大的区域的原始灰度值进行一阶全变分处理，可以去除该区域内的噪声，并且相对于现有技术中的中通滤波、均值滤波以及高斯滤波等方法，可以提高其空间分辨率。而且，通过按照以上公式(2)或(5)来确定像素点的最终灰度值，这可以提高图像的对比度和分辨率。
可以按照如下过程来执行所述调用所构建的第二降噪模型对原始灰度值进行二阶全变分处理的步骤：
当原始图像为二维图像时，可以按照以下公式(7)依次对满足原始灰度值的总变化率小于预设值这个条件的所有区域内的所有像素点的原始灰度值进行二阶梯度处理；针对满足所述条件的每个区域，判断处理后得到的每个像素点的总灰度值是否大于第二预设阈值；针对每个像素点，当判断出处理后得到的总灰度值大于或等于第二预设阈值时，将处理后得到的总灰度值确定为该像素点的最终灰度值，或者按照以下公式(8)来计算该像素点的最终灰度值，而当判断出处理后得到的总灰度值小于第二预设阈值时，按照以下公式(9)对处理后得到的总灰度值进行计算来获得该像素点的最终灰度值。
Figure PCTCN2021118707-appb-000026
Figure PCTCN2021118707-appb-000027
Figure PCTCN2021118707-appb-000028
其中，f x,y、f x-1,y、f x+1,y、f x,y-1和f x,y+1分别表示位于坐标(x,y)、(x-1,y)、(x+1,y)、(x,y-1)和(x,y+1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000029
和
Figure PCTCN2021118707-appb-000030
分别表示经过二阶梯度处理后得到的位于坐标(x,y)处的像素点在x轴和y轴方向 上的灰度值；
Figure PCTCN2021118707-appb-000031
为经过二阶梯度处理后得到的位于坐标(x,y)处的像素点的总灰度值；f′ x,y表示位于坐标(x,y)处的像素点的最终灰度值。
当原始图像为三维图像时，可以按照以下公式(10)依次对满足原始灰度值的总变化率小于预设值这个条件的所有区域内的所有像素点的原始灰度值进行二阶梯度处理；针对满足所述条件的每个区域，判断处理后得到的每个像素点的总灰度值是否大于第二预设阈值；针对每个像素点，当判断出其总灰度值大于或等于第二预设阈值时，将该总灰度值确定为该像素点的最终灰度值，或者按照以下公式(11)来计算该像素点的最终灰度值，而当判断出总灰度值小于第二预设阈值时，按照以下公式(12)对处理后得到的总灰度值进行计算来获得该像素点的最终灰度值。
Figure PCTCN2021118707-appb-000032
Figure PCTCN2021118707-appb-000033
Figure PCTCN2021118707-appb-000034
其中，f x,y,z、f x-1,y,z、f x+1,y-1,z、f x,y-1,z、f x,y+1,z、f x,y,z-1和f x,y,z+1分别表示位于坐标(x,y,z)、(x-1,y,z)、(x+1,y,z)、(x,y-1,z)、(x,y+1,z)(x,y,z-1)、(x,y,z+1)处的像素点的原始灰度值；
Figure PCTCN2021118707-appb-000035
和
Figure PCTCN2021118707-appb-000036
分别表示经过二阶梯度处理后得到的位于坐标(x,y,z)处的像素点在x轴、y轴和z轴方向上的灰度值；
Figure PCTCN2021118707-appb-000037
为经过二阶梯度处理后得到的位于坐标(x,y,z)处的像素点的总灰度值；f′ x,y,z表示位于坐标(x,y,z)处的像素点的最终灰度值。
针对图像的灰度值呈平滑线性变化并且其线性变化的斜率较小的情况，通过调用第二降噪模型对原始灰度值进行二阶全变分处理，可以提高其空间分辨率，并且可以保证图像的真实度。这是因为当原始灰度值呈线性变化时，经过 二阶梯度处理后，像素点在各个方向上的灰度值均为0，从而使得像素点的最终灰度值等于其原始灰度值。另外，通过按照以上公式(8)或(11)来确定像素点的最终灰度值，这可以提高图像的对比度和分辨率。
需要说明的是，上面公式中的坐标(x-1,y)、(x+1,y)、(x,y-1)和(x,y+1)等仅代表与坐标(x,y)的位置关系，其可以不指代具体的坐标数值。例如，位于坐标(x-1,y)和(x+1,y)处的像素点与位于坐标(x,y)处的像素点在x轴方向上相邻，位于坐标(x,y-1)和(x,y+1)处的像素点与位于坐标(x,y)处的像素点在y轴方向上相邻。另外，上面公式中的坐标(x-1,y,z)、(x+1,y,z)、(x,y-1,z)、(x,y+1,z)(x,y,z-1)、(x,y,z+1)也仅代表与坐标(x,y,z)的位置关系。
通过上面描述可以看出，本申请实施例通过将原始图像划分为多个区域，计算每个区域的原始灰度值的总变化率，并且根据每个区域的原始灰度值的总变化率与预设值的大小关系来选择第一降噪模型对原始灰度值进行一阶全变分处理或选择第二降噪模型对原始灰度值进行二阶全变分处理，并且在全变分处理中采用了将一阶梯度处理结果与第一预设阈值比较，或者将二阶梯度结果与第二预设阈值比较，这使得能够去除灰度值低于第一预设阈值或第二预设阈值的位置处的噪声，而在高于第一预设阈值或第二预设阈值的位置会继续保持并增强原来的梯度，使得灰度值高的位置梯度更高，灰度值低的地方梯度更低，也就增强了对比度并提高了分辨率，从而可以实现在去除图像中的噪声的同时提高其空间分辨率的目的。
本申请实施例还提供了一种图像处理装置，如图3所示，其可以包括：
划分单元310，其被配置为按照预设规则将所获取的原始图像划分为多个区域；
计算单元320，其被配置为计算每个区域的原始灰度值的总变化率；
处理单元330，其被配置为判断总变化率是否大于预设值，在判断出原始 灰度值的总变化率大于或等于预设值时，调用所构建的第一降噪模型对原始灰度值进行一阶全变分处理，并且在判断出原始灰度值的总变化率小于预设值时，调用所构建的第二降噪模型对原始灰度值进行二阶全变分处理。
关于上述单元的详细描述，可以参照方法实施例中的对应描述，在此不再赘叙。
通过利用该图像处理装置，可以实现在去除图像中的噪声，并且可以提高其空间分辨率。
图4示出了一个实施例中的计算机设备的结构示意图。该计算机设备具体可以是图1中的终端1000。如图4所示，该计算机设备包括通过系统总线连接的处理器、存储器、网络接口、输入装置和显示器。其中，存储器包括非易失性存储介质和内存储器。该计算机设备的非易失性存储介质可以存储有操作系统，还可以存储有计算机程序，在该计算机程序被处理器执行时，可使得处理器执行上述实施例中描述的图像处理方法。该内存储器中也可储存有计算机程序，该计算机程序被处理器执行时，执行上述实施例中描述的图像处理方法。
图5示出了另一个实施例中的计算机设备的结构示意图。该计算机设备具体可以是图1中的服务器2000。如图5所示，该计算机设备包括通过系统总线连接的处理器、存储器以及网络接口。其中，存储器包括非易失性存储介质和内存储器。该计算机设备的非易失性存储介质存储有操作系统，还可存储有计算机程序，该计算机程序被处理器执行时，可使得处理器执行上述实施例中描述的图像处理方法。该内存储器中也可储存有计算机程序，该计算机程序被处理器执行时，执行上述实施例中描述的图像处理方法。
本领域技术人员可以理解，图4和图5中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件配置。
在一个实施例中，如图6所示，本申请还提供了一种图像处理系统，该图像处理系统可以包括图4或图5中的计算机设备以及与其连接的扫描设备，该扫描设备可以用于通过对目标对象进行扫描而获得原始图像并且将所获得的原始图像提供给计算机设备。该扫描设备可以是能够探测放射性射线的任意设备，例如，可以包括CT扫描仪、MRI扫描仪、PET探测器、PET-CT设备或超声设备等，但不限于此。
在一个实施例中，本申请还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有计算机程序，该计算机程序被执行时能够实现上述方法实施例中描述对应的功能。该计算机程序还可在如图4或图5所示的计算机设备上运行。该计算机设备的存储器包含组成该装置的各个程序模块，各个程序模块构成的计算机程序被执行时能够实现与上述实施例中描述的图像分割方法中的各个步骤所对应的功能。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一非易失性计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储介质、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
上述实施例阐明的系统、设备、装置、单元等，具体可以由半导体芯片、计算机芯片和/或实体实现，或者由具有某种功能的产品来实现。为了描述的方便，描述以上装置时以功能分为各种单元分别描述。当然，在实施本申请时可以把各单元的功能在同一个或多个芯片中实现。
虽然本申请提供了如上述实施例或流程图所述的方法操作步骤，但基于常规或者无需创造性的劳动在所述方法中可以包括更多或者更少的操作步骤。在逻辑性上不存在必要因果关系的步骤中，这些步骤的执行顺序不限于本申请实施例提供的执行顺序。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其它实施例的不同之处。另外，以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
上述实施例是为便于该技术领域的普通技术人员能够理解和使用本申请而描述的。熟悉本领域技术的人员显然可以容易地对这些实施例做出各种修改，并把在此说明的一般原理应用到其它实施例中而不必经过创造性的劳动。因此，本申请不限于上述实施例，本领域技术人员根据本申请的揭示，不脱离本申请范畴所做出的改进和修改都应该在本申请的保护范围之内。

Claims (15)

  1. 一种图像处理方法，其特征在于，所述方法包括：
    按照预设规则将所获取的原始图像划分为多个区域；
    计算每个所述区域的原始灰度值的总变化率；
    判断所述原始灰度值的所述总变化率是否大于预设值，当判断出所述原始灰度值的总变化率大于或等于所述预设值时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理，并且当判断出所述原始灰度值的总变化率小于所述预设值时，调用所构建的第二降噪模型对所述原始灰度值进行二阶全变分处理。
  2. 根据权利要求1所述的方法，其特征在于，按照预设规则将所述原始图像划分为多个区域的步骤包括：
    根据所述原始图像的构成元素、物理属性或方位将所述原始图像划分为多个区域。
  3. 根据权利要求1所述的方法，其特征在于，计算每个所述区域的原始灰度值的总变化率的步骤包括：
    针对每个所述区域，依次计算所述区域内的每两个相邻像素点的原始灰度值之间的差值；
    根据所述原始灰度值的差值与每两个相邻所述像素点的坐标差值来确定两个相邻所述像素点的所述原始灰度值的变化率；
    根据所得到的所述区域内的所有所述像素点的所述原始灰度值的变化率来确定所述区域的所述原始灰度值的总变化率。
  4. 根据权利要求3所述的方法，其特征在于，确定所述区域的所述原始灰度值的总变化率的步骤包括：
    将所述区域内的所有所述像素点的所述原始灰度值的变化率的平均值确定 为所述区域的所述原始灰度值的总变化率；或者
    对所述区域内的所有所述像素点的所述原始灰度值的变化率进行均方差运算以获得所述区域的所述原始灰度值的总变化率。
  5. 根据权利要求1所述的方法，其特征在于，计算每个所述区域的原始灰度值的总变化率的步骤还包括：
    统计每个所述区域内的所有像素点的所述原始灰度值的变化规律，并且当所述原始灰度值呈线性变化时，将所述原始灰度值的线性变化的斜率确定为所述区域的所述原始灰度值的所述总变化率。
  6. 根据权利要求1所述的方法，其特征在于，当所述原始图像为二维图像时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理的步骤包括：
    按照以下公式依次对满足所述原始灰度值的总变化率大于或等于所述预设值的所有区域内的所有像素点的原始灰度值进行一阶梯度处理：
    Figure PCTCN2021118707-appb-100001
    针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大于第一预设阈值；
    针对每个像素点，当判断出处理后得到的所述总灰度值大于或等于所述第一预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按照以下公式来计算所述像素点的最终灰度值：
    f′ x,y＝f x,y+k×[(f x,y-f x-1,y)+(f x,y-f x,y-1)]，
    而当判断出所述总灰度值小于所述第一预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
    f′ x,y＝f x,y-(f x,y-f x-1,y)-(f x,y-f x,y-1)，
    其中，f x，y、f x-1,y和f x,y-1分别表示位于坐标(x,y)、(x-1,y)和(x,y-1)处的像素点的原始灰度值；
    Figure PCTCN2021118707-appb-100002
    表示处理后得到的位于坐标(x,y)处的像素点 的总灰度值；f′ x,y表示位于坐标(x,y)处的像素点的最终灰度值；x和y均为自然数；k为图像增强系数。
  7. 根据权利要求1所述的方法，其特征在于，当所述原始图像为三维图像时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理的步骤包括：
    按照以下公式依次对满足所述原始灰度值的总变化率大于或等于所述预设值的所有区域内的所有像素点的原始灰度值进行一阶梯度处理：
    Figure PCTCN2021118707-appb-100003
    针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大于第一预设阈值；
    针对每个像素点，当判断出处理后得到的所述总灰度值大于或等于所述第一预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按照以下公式来计算所述像素点的最终灰度值：
    f′ x,y,z＝f x,y，z+k×[(f x,y,z-f x-1,y,z )+(f x,y,z-f x,y-1,z )+(f x,y,z-f x,y,z-1)]，
    而当判断出所述总灰度值小于所述第一预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
    f′ x,y,z＝f x,y,z-(f x,y,z-f x-1,y,z)-(f x,y,z-f x,y-1,z)-(f x,y,z-f x,y,z-1)，
    其中，f x,y,z、f x-1,y,z、f x,y-1,z和f x,y,z-1分别表示位于坐标(x,y,z)、(x-1,y,z)、(x,y-1,z)和(x,y,z-1)处的像素点的原始灰度值；
    Figure PCTCN2021118707-appb-100004
    表示处理后得到的位于坐标(x,y,z)处的像素点的总灰度值；f′ x,y,z表示位于坐标(x,y,z)处的像素点的最终灰度值；x、y和z均为自然数；k为图像增强系数。
  8. 根据权利要求1所述的方法，其特征在于，当所述原始图像为二维图像时，调用所述第二降噪模型对所述原始灰度值进行二阶全变分处理的步骤包括：
    按照以下公式依次对满足所述原始灰度值的总变化率小于所述预设值的所有区域内的所有像素点的原始灰度值进行二阶梯度处理：
    Figure PCTCN2021118707-appb-100005
    Figure PCTCN2021118707-appb-100006
    Figure PCTCN2021118707-appb-100007
    针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大于第二预设阈值；
    针对每个所述像素点，当判断出处理后得到的所述总灰度值大于或等于所述第二预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按照以下公式来计算所述像素点的最终灰度值：
    Figure PCTCN2021118707-appb-100008
    而当判断出所述总灰度值小于所述第二预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
    Figure PCTCN2021118707-appb-100009
    其中，f x,y、f x-1,y、f x+1,y、f x,y-1和f x,y+1分别表示位于坐标(x,y)、(x-1,y)、(x+1,y)、(x,y-1)和(x,y+1)处的像素点的原始灰度值；
    Figure PCTCN2021118707-appb-100010
    和
    Figure PCTCN2021118707-appb-100011
    分别表示处理后得到的位于坐标(x,y)处的像素点在x轴和y轴方向上的灰度值；
    Figure PCTCN2021118707-appb-100012
    为处理后得到的位于坐标(x,y)处的像素点的总灰度值；f′ x,y表示位于坐标(x,y)处的像素点的最终灰度值；x和y均为自然数；k为图像增强系数。
  9. 根据权利要求1所述的方法，其特征在于，当所述原始图像为三维图像时，调用所述第二降噪模型对所述原始灰度值进行二阶全变分处理的步骤包括：
    按照以下公式依次对满足所述原始灰度值的总变化率小于所述预设值的所有区域内的所有像素点的原始灰度值进行二阶梯度处理：
    Figure PCTCN2021118707-appb-100013
    Figure PCTCN2021118707-appb-100014
    Figure PCTCN2021118707-appb-100015
    Figure PCTCN2021118707-appb-100016
    针对每个所述区域，判断处理后得到的每个所述像素点的总灰度值是否大 于第二预设阈值；
    针对每个所述像素点，当判断出处理后得到的所述总灰度值大于或等于所述第二预设阈值时，将所述总灰度值确定为所述像素点的最终灰度值，或者按照以下公式来计算所述像素点的最终灰度值：
    Figure PCTCN2021118707-appb-100017
    而当判断出所述总灰度值小于所述第二预设阈值时，按照以下公式对所述总灰度值进行计算来获得所述像素点的最终灰度值：
    Figure PCTCN2021118707-appb-100018
    其中，f x,y,z、f x-1,y,z、f x+1,y-1,z、f x,y-1,z、f x,y+1,z、f x,y,z-1和f x,y,z+1分别表示位于坐标(x,y,z)、(x-1,y,z)、(x+1,y,z)、(x,y-1,z)、(x,y+1,z)(x,y,z-1)、(x,y,z+1)处的像素点的原始灰度值；
    Figure PCTCN2021118707-appb-100019
    和
    Figure PCTCN2021118707-appb-100020
    分别表示处理后得到的位于坐标(x,y,z)处的像素点在x轴、y轴和z轴方向上的灰度值；
    Figure PCTCN2021118707-appb-100021
    为处理后得到的位于坐标(x,y,z)处的像素点的总灰度值；f′ x,y,z表示位于坐标(x,y,z)处的像素点的最终灰度值；x、y和z均为自然数；k为图像增强系数。
  10. 根据权利要求1-9中任一项所述的方法，其特征在于，所述原始图像包括CT图像、MRI图像、PET图像、PET-CT图像或超声图像。
  11. 一种图像处理装置，其特征在于，所述装置包括：
    划分单元，其被配置为按照预设规则将所获取的原始图像划分为多个区域；
    计算单元，其被配置为计算每个所述区域的原始灰度值的总变化率；
    处理单元，其被配置为判断所述总变化率是否大于预设值，在判断出所述原始灰度值的总变化率大于或等于所述预设值时，调用所构建的第一降噪模型对所述原始灰度值进行一阶全变分处理，并且在判断出所述原始灰度值的总变化率小于所述预设值时，调用所构建的第二降噪模型对所述原始灰度值进行二阶全变分处理。
  12. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所 述计算机程序被执行时能够实现如权利要求1-10中任一项所述的图像处理方法。
  13. 一种计算机设备，其特征在于，所述设备包括处理器和存储器，其中，所述存储器上存储有计算机程序，当所述计算机程序被执行时，所述处理器执行如权利要求1-10中任一项所述的图像处理方法。
  14. 一种图像处理系统，其特征在于，所述系统包括如权利要求13中所述的计算机设备以及与所述计算机设备连接的扫描设备。
  15. 根据权利要求14所述的系统，其特征在于，所述扫描设备包括CT扫描仪、MRI扫描仪、PET探测器、PET-CT设备或超声设备。
PCT/CN2021/118707 2020-10-30 2021-09-16 图像处理方法、装置、设备、系统及计算机可读存储介质 WO2022089079A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011193184.XA CN112365413A (zh) 2020-10-30 2020-10-30 图像处理方法、装置、设备、系统及计算机可读存储介质
CN202011193184.X 2020-10-30

Publications (1)

Publication Number Publication Date
WO2022089079A1 true WO2022089079A1 (zh) 2022-05-05

Family

ID=74513316

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/118707 WO2022089079A1 (zh) 2020-10-30 2021-09-16 图像处理方法、装置、设备、系统及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN112365413A (zh)
WO (1) WO2022089079A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116010996A (zh) * 2023-03-01 2023-04-25 上海伯镭智能科技有限公司 一种无人驾驶系统开发数据安全管理方法
CN116485787A (zh) * 2023-06-15 2023-07-25 东莞市立时电子有限公司 一种数据线成型外模外观缺陷检测方法
CN116721099A (zh) * 2023-08-09 2023-09-08 山东奥洛瑞医疗科技有限公司 一种基于聚类的肝脏ct影像的图像分割方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112365413A (zh) * 2020-10-30 2021-02-12 湖北锐世数字医学影像科技有限公司 图像处理方法、装置、设备、系统及计算机可读存储介质
CN113763275A (zh) * 2021-09-09 2021-12-07 深圳市文立科技有限公司 一种自适应图像降噪方法、系统及可读存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105654428A (zh) * 2014-11-14 2016-06-08 联芯科技有限公司 图像降噪方法及系统
CN109584185A (zh) * 2018-12-19 2019-04-05 深圳市华星光电半导体显示技术有限公司 图像处理方法
CN110689496A (zh) * 2019-09-25 2020-01-14 北京迈格威科技有限公司 降噪模型的确定方法、装置、电子设备和计算机存储介质
CN112365413A (zh) * 2020-10-30 2021-02-12 湖北锐世数字医学影像科技有限公司 图像处理方法、装置、设备、系统及计算机可读存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103942755B (zh) * 2013-01-23 2017-11-17 深圳市腾讯计算机系统有限公司 图像亮度调节方法和装置
US10297024B2 (en) * 2015-09-30 2019-05-21 Shanghai United Imaging Healthcare Co., Ltd. System and method for determining a breast region in a medical image
CN111183630B (zh) * 2017-08-02 2021-08-10 深圳传音通讯有限公司 一种智能终端的照片处理方法及处理装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105654428A (zh) * 2014-11-14 2016-06-08 联芯科技有限公司 图像降噪方法及系统
CN109584185A (zh) * 2018-12-19 2019-04-05 深圳市华星光电半导体显示技术有限公司 图像处理方法
CN110689496A (zh) * 2019-09-25 2020-01-14 北京迈格威科技有限公司 降噪模型的确定方法、装置、电子设备和计算机存储介质
CN112365413A (zh) * 2020-10-30 2021-02-12 湖北锐世数字医学影像科技有限公司 图像处理方法、装置、设备、系统及计算机可读存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WANG, YUANQUAN ET AL.: "MTV: Modified total variation model for image noise removal", ELECTRONICS LETTERS, vol. 47, no. 10, 12 May 2011 (2011-05-12), XP006038873, DOI: 10.1049/EL:20103505 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116010996A (zh) * 2023-03-01 2023-04-25 上海伯镭智能科技有限公司 一种无人驾驶系统开发数据安全管理方法
CN116010996B (zh) * 2023-03-01 2023-06-13 上海伯镭智能科技有限公司 一种无人驾驶系统开发数据安全管理方法
CN116485787A (zh) * 2023-06-15 2023-07-25 东莞市立时电子有限公司 一种数据线成型外模外观缺陷检测方法
CN116485787B (zh) * 2023-06-15 2023-08-22 东莞市立时电子有限公司 一种数据线成型外模外观缺陷检测方法
CN116721099A (zh) * 2023-08-09 2023-09-08 山东奥洛瑞医疗科技有限公司 一种基于聚类的肝脏ct影像的图像分割方法
CN116721099B (zh) * 2023-08-09 2023-11-21 山东奥洛瑞医疗科技有限公司 一种基于聚类的肝脏ct影像的图像分割方法

Also Published As

Publication number Publication date
CN112365413A (zh) 2021-02-12

Similar Documents

Publication Publication Date Title
WO2022089079A1 (zh) 图像处理方法、装置、设备、系统及计算机可读存储介质
CN109523584B (zh) 图像处理方法、装置、多模态成像系统、存储介质及设备
JP6362333B2 (ja) 画像処理装置、画像処理方法、およびプログラム
WO2021135773A1 (zh) 图像重建方法、装置、设备、系统及计算机可读存储介质
WO2020125498A1 (zh) 心脏磁共振图像分割方法、装置、终端设备及存储介质
WO2022089221A1 (zh) 医学图像分割方法、装置、设备、系统及计算机存储介质
US7929746B2 (en) System and method for processing imaging data
US9058650B2 (en) Methods, apparatuses, and computer program products for identifying a region of interest within a mammogram image
CN109544657B (zh) 医学图像迭代重建方法、装置、计算机设备和存储介质
JP6422746B2 (ja) 核医学診断装置、医用画像処理装置、ソース位置の推定方法およびソース位置推定プログラム
WO2024066049A1 (zh) 一种pet图像去噪的方法、终端设备及可读存储介质
US20230377314A1 (en) Out-of-distribution detection of input instances to a model
EP3067863A1 (en) Image noise reduction
WO2021082819A1 (zh) 一种图像生成方法、装置及电子设备
CN113034508B (zh) 超声图像病灶的分割方法、装置和计算机设备
CN114549594A (zh) 图像配准方法、装置和电子设备
Yilmaz et al. Noise removal of CBCT images using an adaptive anisotropic diffusion filter
CN114299081B (zh) 一种上颌窦cbct图像分割方法、装置、存储介质及电子设备
JP6625165B2 (ja) 画像処理装置、画像処理方法、およびプログラム
CN112837317A (zh) 基于乳腺超声图像增强的病灶分类方法、装置及存储介质
WO2022247004A1 (zh) 图像预测模型的训练及应用方法、装置、设备、存储介质
CN114298934B (zh) 一种基于像素调节的面颊夹显影弱化方法、装置
CN111161239B (zh) 医学图像分析方法、装置、存储介质及计算机设备
CN116309922B (zh) Ct灌注影像的去伪影方法、装置、设备和存储介质
CN114255288A (zh) 心脏图像重建方法、装置、计算机设备和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21884809

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21884809

Country of ref document: EP

Kind code of ref document: A1