WO2013146283A1 - 画像処理装置及び画像処理方法 - Google Patents
画像処理装置及び画像処理方法 Download PDFInfo
- Publication number
- WO2013146283A1 WO2013146283A1 PCT/JP2013/057118 JP2013057118W WO2013146283A1 WO 2013146283 A1 WO2013146283 A1 WO 2013146283A1 JP 2013057118 W JP2013057118 W JP 2013057118W WO 2013146283 A1 WO2013146283 A1 WO 2013146283A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- region
- image processing
- feature amount
- smoothed
- Prior art date
Links
- 238000012545 processing Methods 0.000 title claims abstract description 33
- 238000003672 processing method Methods 0.000 title claims description 15
- 238000004364 calculation method Methods 0.000 claims description 34
- 230000009467 reduction Effects 0.000 claims description 29
- 239000000284 extract Substances 0.000 claims description 9
- 238000013459 approach Methods 0.000 claims description 5
- 230000007423 decrease Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims 6
- 238000012886 linear function Methods 0.000 abstract description 5
- 238000000034 method Methods 0.000 description 23
- 238000009499 grossing Methods 0.000 description 8
- 238000005452 bending Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 210000000056 organ Anatomy 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 230000007480 spreading Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/032—Transmission computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5205—Devices using data or image processing specially adapted for radiation diagnosis involving processing of raw data to produce diagnostic data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/008—Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20224—Image subtraction
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2211/00—Image generation
- G06T2211/40—Computed tomography
Definitions
- the present invention relates to an image processing apparatus that performs image processing on an image obtained by an X-ray CT apparatus or the like, and relates to a technique for removing streaky artifacts while maintaining the edge of a structure.
- the X-ray CT apparatus reconstructs a tomographic image of a subject by irradiating X-rays from around the subject and backprojecting actual projection data acquired at a plurality of projection angles.
- Non-Patent Document 1 and Non-Patent Document 2 propose a method of reducing streak-like artifacts by applying the above-described image reconstruction method after smoothing projection data.
- a method for reducing these streak-like artifacts is referred to as a “projection data smoothing filter”.
- the image obtained by the projection data smoothing filter is blurred at the edge of a relatively high-contrast structure accompanying the streak-like artifact reduction effect.
- the effect of reducing streak-like artifacts and the effect of preserving the edge of the structure are in a trade-off relationship, and the latter effect decreases as the former effect increases. Therefore, in the case of reducing strong streak-like artifacts, large blurring occurs at the edges of the structure. Since the edge of the structure is not preferable in clinical diagnosis, an image in which the edge of the structure is maintained and the streak-like artifact is removed (hereinafter referred to as “target image”) is desired.
- an image obtained from the projection data by the above-described image reconstruction method (hereinafter referred to as “original image”) and the above-described image reconstruction method from the projection data to which the projection data smoothing filter is applied.
- original image an image obtained from the projection data by the above-described image reconstruction method
- smoothed images images obtained by the above
- weighted addition is performed so that the ratio of the pixel value of the smoothed image is increased at the pixel portion where the streak-like artifact is generated on the image and the ratio of the pixel value of the original image is increased at the edge portion of the structure.
- a low-resolution image having a lower resolution than the original image is created from the original image, an edge is detected from the low-resolution image, and a low-resolution edge image is created.
- a weighting coefficient is calculated from the edge luminance value by a non-linear function.
- the difference between the maximum value and the minimum value of the pixels in a predetermined region including the target pixel, the variance value, the density gradient, and the like are used for edge detection.
- the edge of the structure is emphasized by weighting and adding the image obtained by subtracting the low-resolution image from the original image using the weighting coefficient described above.
- the image is divided into small areas, the standard deviation value of the small area is calculated, a histogram with the standard deviation value as a class is created for all small areas, and the discrimination class value is determined from the histogram. And each small region is classified into a flat region and a non-flat region. Detecting the edge of the structure on the image is equivalent to identifying flat and non-flat areas. If both can be classified, the pixel value of the original image is used in the non-flat region, and the pixel value of the smoothed image is used in the flat region, so that the weighted addition of the original image and the smoothed image can be performed.
- the strong streak-like artifact has a characteristic close to a linear structure unlike the granular noise generated in the image. Then, as in the method described in Patent Document 1, when performing edge detection using the difference between the maximum value and the minimum value of the pixel value of the predetermined region including the target pixel, the variance value, the density gradient, A strong streak-like artifact is likely to be detected as an edge.
- Patent Document 1 and Patent Document 2 are intended to classify noise and structures in a uniform region, and streak-like artifacts. This is because the index for classifying the structure is not used.
- the present invention has been made in view of the above-described problems, and an object of the present invention is to generate a target image in which the edge of a structure is maintained and streak-like artifacts are removed.
- An image processing apparatus is provided.
- an original image generated from original information and corresponding smoothed images generated from the same original information and at least streak-like artifacts are reduced.
- An image processing apparatus for generating a target image in which edges of a structure are maintained and streak-like artifacts are removed by weight-adding pixels using a weighting coefficient, wherein the feature amounts of the original image and the smoothed image An image processing apparatus comprising: a function shape determining unit that determines a shape of a nonlinear function based on the first step; and a weighting factor calculating unit that calculates the weighting factor based on the nonlinear function.
- the second invention weights and adds the corresponding pixels of the original image generated from the original information and the smoothed image generated from the same original information and having at least streak-like artifacts reduced by a weighting coefficient.
- An image processing method comprising: a function shape determining step for determining; and a weighting coefficient calculating step for calculating the weighting coefficient based on the nonlinear function.
- an image processing apparatus capable of generating a target image in which the edge of a structure is maintained and streak-like artifacts are removed.
- the X-ray CT apparatus 1 performs processing of data obtained from the scanner 2 on which the X-ray tube 11 and the detector 12 are mounted, the bed 4 on which the subject 10 is placed, and the detector 12.
- An arithmetic device 5 an input device 6 such as a mouse, a trackball, a keyboard, and a touch panel, and a display device 7 for displaying a reconstructed image and the like are included.
- the operator inputs shooting conditions and reconstruction parameters via the input device 6.
- the imaging conditions are, for example, a bed feeding speed, a tube current, a tube voltage, and a slice position.
- the reconstruction parameter is, for example, a region of interest, a reconstructed image size, a backprojection phase width, a reconstruction filter function, or the like.
- the X-ray CT apparatus 1 is roughly composed of a scanner 2, an operation unit 3, and a bed 4.
- the scanner 2 includes an X-ray tube 11 (X-ray source), a detector 12 (X-ray detector), a collimator 13, a drive unit 14, a central control unit 15, an X-ray control unit 16, a high voltage generation unit 17, and a scanner control.
- the device 18 includes a bed control device 19, a bed movement measurement device 20, a collimator control device 21, a preamplifier 22, an A / D converter 23, and the like.
- the X-ray CT apparatus 1 is a multi-slice CT that uses a detector 12 in which detector elements are arranged in a two-dimensional direction, and a single that uses a detector 12 in which detector elements are arranged in one row, that is, in a one-dimensional direction (channel direction only). Broadly divided into slice CT.
- multi-slice CT an X-ray beam spreading in a cone shape or a pyramid shape is irradiated from an X-ray tube 11 as an X-ray source in accordance with the detector 12.
- an X-ray beam spreading in a fan shape is emitted from the X-ray tube 11.
- X-ray irradiation is performed while the gantry section circulates around the subject 10 placed on the bed 4.
- the central controller 15 inputs imaging conditions and reconstruction parameters from the input device 6 in the operation unit 3, and sends control signals necessary for imaging to the collimator controller 21, the X-ray controller 16, the scanner controller 18, and the bed control. Transmit to device 19.
- the collimator control device 21 controls the position of the collimator 13 based on the control signal.
- the X-ray control device 16 controls the high voltage generator 17 based on the control signal.
- the high voltage generator 17 applies a tube voltage and a tube current to the X-ray tube 11.
- electrons with energy corresponding to the applied tube voltage are emitted from the cathode, and the emitted electrons collide with the target (anode), whereby X-rays with energy corresponding to the electron energy are Is irradiated.
- the scanner control device 18 controls the drive device 14 based on the control signal.
- the driving device 14 circulates around the subject 10 around a gantry portion on which the X-ray tube 11, the detector 12, the preamplifier 22, and the like are mounted.
- the bed control device 19 controls the bed 4 based on the control signal.
- the X-ray irradiated from the X-ray tube 11 is limited in the irradiation region by the collimator 13, is absorbed (attenuated) according to the X-ray attenuation coefficient in each tissue in the subject 10, passes through the subject 10, It is detected by a detector 12 arranged at a position facing the tube 11.
- the detector 12 includes a plurality of detection elements arranged in a two-dimensional direction (a channel direction and a column direction perpendicular to the channel direction). X-rays received by each detection element are converted into actual projection data.
- the X-rays detected by the detector 12 are converted into current, amplified by the preamplifier 22, converted into digital data by the A / D converter 23, LOG converted, calibrated, and used as actual projection data. Input to the arithmetic unit 5.
- the computing device 5 includes a reconstruction computing device 31, an image processing device 32, and the like.
- the input / output device 9 includes an input device 6, a display device 7, a storage device 8 (storage unit), and the like.
- the reconstruction calculation device 31 performs an image reconstruction process using the actual projection data, and generates a reconstructed image.
- the reconstruction calculation device 31 superimposes the reconstruction filter on the actual projection data of each view to generate filter-corrected projection data, and performs back-projection processing with weighting the filter-corrected projection data in the view direction.
- a tomographic image is imaged non-destructively as a distribution map of the X-ray attenuation coefficient inside the subject 10.
- the reconstruction calculation device 31 stores the generated reconstruction image in the storage device 8. Further, the reconstruction calculation device 31 displays the CT image on the display device 7. Alternatively, the image processing device 32 performs image processing on the reconstructed image stored in the storage device 8 and displays it on the display device 7 as a CT image.
- each pixel corresponding to the original image generated from the original information acquired by the X-ray CT apparatus 1 and the like and the smoothed image generated from the same original information and at least streak-like artifacts are reduced.
- An image processing method for generating a target image in which the edge of the structure is maintained and streak-like artifacts are removed by weighted addition using a weighting coefficient will be described.
- the image processing method of the present invention generates a target image so as to satisfy the following three constraints.
- the pixel value of the target image is within the range of the pixel value of the original image and the pixel value of the smoothed image related to the corresponding pixel.
- the target image is a smooth image spatially (that is, in one image space).
- the constraint condition 1 is the original information acquired by the X-ray CT apparatus 1 (excluding noise). ) Is a condition for not losing.
- the constraint condition 1 is min (pixel value of the original image, pixel value of the smoothed image) ⁇ pixel value of the target image ⁇ max (pixel value of the original image, pixel value of the smoothed image).
- Min () is an operator that outputs a minimum value
- max () is an operator that outputs a maximum value.
- the constraint condition 2 is a condition for avoiding discontinuity that may occur in the target image.
- the constraint condition 3 is a condition for mainly removing graininess noise and streak-like artifacts.
- the image processing method of the present invention is particularly characterized in that the constraint condition 3 is satisfied.
- the weighting coefficient used for weighted addition is determined by the state coefficient representing the state of the original image and the smoothed image.
- the state coefficient is a value of a nonlinear function whose variable is the pixel value difference between the target pixel of the original image or the smoothed image and its neighboring pixels.
- the nonlinear function is a smooth continuous function, has a plurality of arbitrary parameters, and can be arbitrarily adjusted in shape, and is not particularly limited. If the non-linear function is a smooth continuous function, it satisfies the above-mentioned “(constraint condition 2) that the target image is a smooth image spatially (that is, in one image space)”. Another example of the nonlinear function is a logistic function.
- a state coefficient g s (x) of the target pixel s is calculated by a generalized Gaussian function. It is defined by the following formula.
- p is an arbitrary parameter for adjusting the slope of the generalized Gaussian function, and is set to the same value for all slices.
- ⁇ is an arbitrary parameter for adjusting the bending position of the generalized Gaussian function.
- w sr is a weighting coefficient corresponding to the distance between the target pixel s and the neighboring pixel r, and is defined as, for example, the following equation.
- l sr is the distance between the target pixel s and the neighboring pixel r.
- L is the ratio of the pixel size to the detector element size.
- w sr need not be limited to Equation (2), and may be a function that increases in value as the adjacent pixel r is closer to the target pixel s.
- the slope p of the set N s, and generalized Gaussian function of neighboring pixels of the pixel of interest s shall be determined empirically.
- FIG. 3 illustrates changes in the function shape when ⁇ is set to arbitrary constants c, 2c, and 4c with respect to the pixel value difference between the target pixel s and the neighboring pixel r. If the pixel value difference is fixed, the state coefficient is likely to be close to 1 as ⁇ is increased.
- the state coefficient is a value close to approximately 1.
- a parameter ⁇ for adjusting the bending position of the function is determined.
- the computing device 5 (hereinafter referred to as “computing device 5”) of the X-ray CT apparatus 1 determines the shape of the nonlinear function based on the feature amounts of the original image and the smoothed image. The process for determining the shape of the nonlinear function will be described with reference to FIG.
- the arithmetic device 5 divides the area of the original image and the smoothed image into small areas corresponding to each other. For example, as shown in FIG. 6, the arithmetic device 5 divides the regions of the original image 41 and the smoothed image 42 into a grid and divides it into small regions. The size of the small area is determined empirically.
- the method for dividing the small area is not limited to this example.
- the area of the original image 41 and the smoothed image 42 may be divided into small areas by the same division method.
- the shape of the small area is not limited to a rectangle.
- a plurality of adjacent pixels may be included in the same small area. Further, one pixel may be included in a plurality of small regions.
- the arithmetic device 5 calculates the variation value of the original image 41 and the smoothed image 42 from the pixel values of the pixels included in the small region for each small region related to both the original image 41 and the smoothed image 42.
- the variation value is, for example, a standard deviation value or (maximum value ⁇ minimum value).
- the variation value is a value indicating variation in each small region, and may be a statistic calculated from the pixel values of the pixels included in each small region. In the following, in order to avoid confusion, the standard deviation value will be described as an example.
- the arithmetic device 5 calculates the variation value reduction rate of the smoothed image 42 based on the variation value of the original image 41 for each small region.
- the reduction rate of the standard deviation value is calculated.
- the reduction rate of the standard deviation value of the smoothed image 42 based on the standard deviation value of the original image 41 is set as ⁇ i and is calculated by the following equation.
- the arithmetic device 5 calculates the standard deviation value reduction rate ⁇ i in all the small regions based on the equation (3).
- equation (3) the shape of the nonlinear function is determined using the feature amount (for example, the reduction rate ⁇ i ) calculated from the pixel values of both the original image 41 and the smoothed image 42, and is shown in FIG.
- Patent Documents 1 and 2 related to a method of mixing two images having the same original information
- Patent Documents 1 and 2 related to a method of mixing two images having the same original information
- only feature values calculated from pixel values of pixels included in only one image are used.
- a standard deviation value calculated from pixel values of pixels included in only one image it is impossible to distinguish whether noise has been reduced or a structure has been blurred by an arbitrary smoothing process. Therefore, the conventional technique cannot generate a target image from which streak-like artifacts are removed as in the present invention.
- the arithmetic device 5 extracts a feature amount calculation region from a set of small regions based on the reduction rate ⁇ i .
- the arithmetic device 5 extracts a small region having the maximum reduction rate as a feature amount calculation region from small regions included in a predetermined range (for example, in the same slice).
- the predetermined range may be within a plurality of slices or all slices.
- the arithmetic device 5 determines the shape of the nonlinear function from the feature amount calculated from the pixel value of the feature amount calculation region (kth small region).
- the feature amount is calculated using the standard deviation value ⁇ k (org) of the original image 41 calculated in step S202 and the standard deviation value ⁇ k (smt) of the smoothed image 42 calculated in step S203.
- ⁇ k (org) the standard deviation value of the original image 41 calculated in step S202
- ⁇ k (smt) of the smoothed image 42 calculated in step S203.
- Arbitrary constant is set to ⁇ (0 ⁇ ⁇ 1), and when the pixel value difference is ⁇ k (generic symbol of ⁇ k (org) and ⁇ k (smt)) , the state coefficient takes the value of ⁇ .
- ⁇ is a real value such as 0.99, 0.98, 0.97, for example.
- the parameter ⁇ for adjusting the bending position of the function is defined as the following expression.
- Equation (4) parameters for adjusting the bending position of the nonlinear function calculated in each of the original image 41 and the smoothed image 42 are set as ⁇ k (org) and ⁇ k (smt) .
- the arithmetic device 5 dynamically determines the shape of the nonlinear function for each predetermined range (for example, for each slice) by the processing shown in FIG.
- One of the features in the first embodiment is that a region where the reduction rate ⁇ i of the standard deviation value of the smoothed image 42 with respect to the original image 41 is maximized is extracted as a feature amount calculation region, and the standard deviation in the feature amount calculation region is extracted. The point is to determine the shape of the nonlinear function using the value.
- the projection data smoothing filter described above has a characteristic of greatly reducing streak-like artifacts. From this characteristic, the standard deviation value reduction rate ⁇ i becomes large in a small region including streak-like artifacts. Further, the projection data smoothing filter has a characteristic that it is accompanied by blurring of the structure. From this characteristic, the standard deviation value of the original image increases in a small region including a structure, and the denominator of Equation (3) increases, resulting in a reduction in standard deviation value reduction rate ⁇ i . That is, there is a high possibility that streak-like artifacts are included in the region where the standard deviation value reduction rate ⁇ i is maximum.
- Step S102> The arithmetic device 5 calculates the state coefficients of the original image 41 and the smoothed image 42 using a nonlinear function whose shape is determined in step S101.
- the processing for calculating the state coefficient follows equation (1).
- the arithmetic device 5 calculates a weighting coefficient for each pixel of the original image 41 and the smoothed image 42 using the state coefficient calculated in step S102.
- the weighting coefficient related to the target pixel s is set to ⁇ s .
- the calculation process of ⁇ s follows any of the following formula (5), formula (6), or formula (7).
- weighting coefficient values of the corresponding pixels are calculated using the state coefficients of both the original image 41 and the smoothed image 42 as shown in the equations (5) to (7). It is a point to do.
- Equation (5) and Equation (7) are obtained when the state coefficient is a value close to 0 in at least one of the original image 41 and the smoothed image 42, and it is determined that the target pixel s corresponds to the edge portion of the structure.
- the weighting coefficient takes a value close to zero.
- both the original image 41 and the smoothed image 42 are weighted when it is determined that the state coefficient has a value close to 1 and the target pixel s corresponds to a flat region (for example, a region considered to be the same organ). The coefficient is close to 1.
- Equation (6) indicates that the weighting coefficient is close to 0 when the state coefficients of both the original image 41 and the smoothed image 42 are close to 0. Further, in the equation (6), when the state coefficients of both the original image 41 and the smoothed image 42 are close to 1, the weighting coefficient is close to 1. On the other hand, in the equation (6), when the difference between the values of the state coefficients of the original image 41 and the smoothed image 42 is large, the average value of both is the weighting coefficient.
- Equation (5) or Equation (7) when it is desired to sharply restore the edge of the structure in the target image. Further, when it is desired to smoothly restore the edge of the structure in the target image, it is desirable to use Equation (6). Both are used properly depending on the application.
- Equation (8) By performing weighted addition using Equation (8), the above-mentioned ⁇ (Constraint 1)
- the pixel value of the target image is within the range of the pixel value of the original image and the pixel value of the smoothed image related to the corresponding pixel. To meet.
- FIG. 7 shows a schematic diagram of an original image 41, a smoothed image 42, and a target image 43.
- streak-like artifacts 51 are generated in the original image 41.
- the smoothed image 42 the edge blur 52 of the structure occurs in the smoothed image 42.
- a target image 43 in which the edge of the structure is maintained and the streak-like artifact 51 is removed is generated from the original image 41 and the smoothed image 42. can do.
- the edge of the structure is maintained and the streak-like artifact 51 is removed.
- a small region in which the reduction rate of the variation value of the smoothed image 42 with respect to the original image 41 is maximized is extracted as the feature amount calculation region, and the nonlinear function is calculated based on the feature amount of the feature amount calculation region.
- a parameter ⁇ for adjusting the bending position is determined.
- the relationship between the variation value reduction rate and the variation value of the original image 41 or the smoothed image 42 is analyzed in more detail, and a feature amount calculation region is extracted. This improves the accuracy of extracting a flat region including the streak-like artifact 51 as a feature amount calculation region.
- the computing device 5 is the upper m of the standard deviation value reduction rate ⁇ i in a small area included in a predetermined range (for example, in the same slice). Extract regions. That is, the arithmetic unit 5 extracts from the small area where the standard deviation value reduction rate ⁇ i is the largest to the m-th order small area.
- m is an arbitrary constant and is determined empirically.
- the arithmetic unit 5 extracts, for each of the original image 41 and the smoothed image 42, a small region having the maximum standard deviation value among the upper m regions as a feature amount calculation region. Subsequent processing, such as feature amount calculation processing, nonlinear function shape determination processing, state coefficient calculation processing, weighting coefficient calculation processing, weighted addition processing, and the like are the same as those in the first embodiment.
- the accuracy of extracting a flat region including streak-like artifacts 51 as a feature amount calculation region is improved.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Optics & Photonics (AREA)
- Veterinary Medicine (AREA)
- Radiology & Medical Imaging (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- High Energy & Nuclear Physics (AREA)
- Pulmonology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
Description
)を失わないための条件である。制約条件1を言い換えると、min(原画像の画素値、平滑化画像の画素値)≦目的画像の画素値≦max(原画像の画素値、平滑化画像の画素値)、となる。なお、min()は最小値を出力する演算子であり、max()は最大値を出力する演算子である。
本発明の画像処理方法は、特に、制約条件3を満たすことが特徴である。
以降では、図4に示す処理フローの流れに沿って、適宜他の図を参照しながら、本発明の第1の実施形態について説明する。
X線CT装置1の演算装置5(以降、「演算装置5」と表記する。)は、原画像と平滑化画像の特徴量に基づき、非線形関数の形状を決定する。非線形関数の形状を決定する処理については、図5を参照しながら説明する。
演算装置5は、原画像及び平滑化画像の領域を、互いに対応する小領域に分割する。例えば、図6に示すように、演算装置5は、原画像41及び平滑化画像42の領域を格子状に区切り、小領域に分割する。小領域の大きさは、経験的に決める。
演算装置5は、原画像41及び平滑化画像42の両方に係る各小領域に対して、小領域に含まれる画素の画素値から、原画像41及び平滑化画像42のばらつき値を算出する。
ばらつき値は、例えば、標準偏差値や、(最大値-最小値)などである。ばらつき値は、各小領域のばらつきを示す値であって、各小領域に含まれる画素の画素値から算出される統計量であれば良い。以降では、混乱を避ける為に、標準偏差値を例にして説明する。
演算装置5は、小領域ごとに、原画像41のばらつき値を基準とする平滑化画像42のばらつき値の低減率を算出する。本実施形態では、標準偏差値の低減率を算出する。i番目の小領域について、原画像41の標準偏差値を基準とする平滑化画像42の標準偏差値の低減率をρiとおき、次式で算出する。
演算装置5は、低減率ρiに基づいて、小領域の集合から特徴量算出領域を抽出する。
第1の実施の形態では、演算装置5は、所定の範囲内(例えば、同一スライス内)に含まれる小領域の中で、低減率が最大の小領域を、特徴量算出領域として抽出する。尚、所定の範囲は、複数のスライス内としても良いし、全てのスライス内としても良い。
演算装置5は、特徴量算出領域(k番目の小領域)の画素値から算出される特徴量から、非線形関数の形状を決定する。以下では、ステップS202において算出される原画像41の標準偏差値σk (org)及びステップS203において算出される平滑化画像42の標準偏差値σk (smt)を用いて、特徴量を算出する例を説明する。但し、本発明は、この例に限定されるわけではなく、標準偏差値に代えて、他のばらつき値(例えば、小領域内の画素の最大値および最小値の差)を用いても良い。
演算装置5は、ステップS101において形状が決定される非線形関数によって、原画像41及び平滑化画像42の状態係数を算出する。状態係数の算出処理は、式(1)に従う。
演算装置5は、ステップS102において算出される状態係数を用いて、原画像41及び平滑化画像42の各画素について加重係数を算出する。原画像41の画像ベクトルをx(org)={x1 (org),・・・,xJ (org)}とおき、平滑化画像42の画像ベクトルをx(smt)={x1 (smt),・・・,xJ (smt)}とおく。画像の加重加算において、注目画素sに係る加重係数をλsとおく。λsの算出処理は、以下に示す式(5)、式(6)又は式(7)のいずれかに従う。
演算装置5は、ステップS103において算出される加重係数を用いて、原画像41と平滑化画像42の加重加算を行い、目的画像を生成する。加重加算後の目的画像の画像ベクトルをx(mrg)={x1 (mrg),・・・,xJ (mrg)}とする。演算装置5は、目的画像の注目画素sについて、次式を用いて加重加算を行う。
また、平滑化画像42では、構造物のエッジの暈け52が発生している。図4に示す本発明の画像処理方法によれば、このような原画像41及び平滑化画像42から、構造物のエッジが保たれ、かつストリーク状のアーチファクト51が除去される目的画像43を生成することができる。図7に示すように、目的画像43では、構造物のエッジが保たれており、かつストリーク状のアーチファクト51が除去されている。
以降では、本発明における第2の実施形態について説明する。尚、第1の実施形態と共通する内容については説明を省略する。
Claims (14)
- 原情報から生成される原画像、及び同一の前記原情報から生成され、少なくともストリーク状のアーチファクトが低減されている平滑化画像の対応する各画素を加重係数によって加重加算して、構造物のエッジが保たれ、かつストリーク状のアーチファクトが除去される目的画像を生成する画像処理装置であって、
前記原画像及び前記平滑化画像の特徴量に基づき、非線形関数の形状を決定する関数形状決定部と、
前記非線形関数に基づいて前記加重係数を算出する加重係数算出部と、
を具備することを特徴とする画像処理装置。 - 前記関数形状決定部は、
前記原画像及び前記平滑化画像の領域を、互いに対応する小領域に分割する分割部と、
前記原画像及び前記平滑化画像の両方に係る前記小領域に対してばらつき値を算出し、前記小領域ごとに前記原画像のばらつき値を基準とする前記平滑化画像のばらつき値の低減率を算出する低減率算出部と、
前記低減率に基づいて、前記小領域の集合から特徴量算出領域を抽出する領域抽出部と、
前記特徴量算出領域の画素値から、前記特徴量を算出する特徴量算出部と、
を含むことを特徴とする請求項1に記載の画像処理装置。 - 前記領域抽出部は、所定の範囲において前記低減率が最大の前記小領域を、前記特徴量算出領域として抽出する
ことを特徴とする請求項2に記載の画像処理装置。 - 前記領域抽出部は、所定の範囲において前記低減率が最大の前記小領域から一定順位の前記小領域までの中で、前記ばらつき値が最大の前記小領域を、前記特徴量算出領域として抽出する
ことを特徴とする請求項2に記載の画像処理装置。 - 前記非線形関数は、前記原画像又は前記平滑化画像の注目画素とその近接画素との画素値差を変数とする滑らかな連続関数であることを特徴とする請求項1に記載の画像処理装置。
- 前記非線形関数は、前記画素値差が小さいほど1に近づき、前記画素値が大きいほど0に近づく関数であることを特徴とする請求項5に記載の画像処理装置。
- 前記加重係数算出部は、前記非線形関数を用いて前記原画像及び前記平滑化画像のそれぞれの状態を表す状態係数を算出し、前記状態係数に基づいて前記加重係数を算出することを特徴とする請求項1に記載の画像処理装置。
- 原情報から生成される原画像、及び同一の前記原情報から生成され、少なくともストリーク状のアーチファクトが低減されている平滑化画像の対応する各画素を加重係数によって加重加算して、構造物のエッジが保たれ、かつストリーク状のアーチファクトが除去される目的画像を生成する画像処理方法であって、
前記原画像及び前記平滑化画像の特徴量に基づき、非線形関数の形状を決定する関数形状決定ステップと、
前記非線形関数に基づいて前記加重係数を算出する加重係数算出ステップと、
を含むことを特徴とする画像処理方法。 - 前記関数形状決定ステップは、
前記原画像及び前記平滑化画像の領域を、互いに対応する小領域に分割する分割ステップと、
前記原画像及び前記平滑化画像の両方に係る前記小領域に対してばらつき値を算出し、前記小領域ごとに前記原画像のばらつき値を基準とする前記平滑化画像のばらつき値の低減率を算出する低減率算出ステップと、
前記低減率に基づいて、前記小領域の集合から特徴量算出領域を抽出する領域抽出ステップと、
前記特徴量算出領域の画素値から、前記特徴量を算出する特徴量算出ステップと、
を含むことを特徴とする請求項8に記載の画像処理方法。 - 前記領域抽出ステップは、所定の範囲において前記低減率が最大の前記小領域を、前記特徴量算出領域として抽出する
ことを特徴とする請求項9に記載の画像処理方法。 - 前記領域抽出ステップは、所定の範囲において前記低減率が最大の前記小領域から一定順位の前記小領域までの中で、前記ばらつき値が最大の前記小領域を、前記特徴量算出領域として抽出する
ことを特徴とする請求項9に記載の画像処理方法。 - 前記非線形関数は、前記原画像又は前記平滑化画像の注目画素とその近接画素との画素値差を変数とする滑らかな連続関数であることを特徴とする請求項8に記載の画像処理方法。
- 前記非線形関数は、前記画素値差が小さいほど1に近づき、前記画素値が大きいほど0に近づく関数であることを特徴とする請求項12に記載の画像処理方法。
- 前記加重係数算出ステップは、前記非線形関数を用いて前記原画像及び前記平滑化画像のそれぞれの状態を表す状態係数を算出し、前記状態係数に基づいて前記加重係数を算出することを特徴とする請求項8に記載の画像処理方法。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/372,330 US9454804B2 (en) | 2012-03-27 | 2013-03-14 | Image processing device and image processing method |
IN5834DEN2014 IN2014DN05834A (ja) | 2012-03-27 | 2013-03-14 | |
CN201380005062.5A CN104066378B (zh) | 2012-03-27 | 2013-03-14 | 图像处理装置以及图像处理方法 |
JP2014507658A JP6124868B2 (ja) | 2012-03-27 | 2013-03-14 | 画像処理装置及び画像処理方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012-070823 | 2012-03-27 | ||
JP2012070823 | 2012-03-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013146283A1 true WO2013146283A1 (ja) | 2013-10-03 |
Family
ID=49259547
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/057118 WO2013146283A1 (ja) | 2012-03-27 | 2013-03-14 | 画像処理装置及び画像処理方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US9454804B2 (ja) |
JP (1) | JP6124868B2 (ja) |
CN (1) | CN104066378B (ja) |
IN (1) | IN2014DN05834A (ja) |
WO (1) | WO2013146283A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2015100539A (ja) * | 2013-11-26 | 2015-06-04 | 日立アロカメディカル株式会社 | 超音波診断装置 |
US10217248B2 (en) | 2014-08-27 | 2019-02-26 | General Electric Company | Method for removing streak from detector cell with performance difference |
JP2020078377A (ja) * | 2018-11-12 | 2020-05-28 | 株式会社日立製作所 | 画像再構成装置および画像再構成方法 |
JP2020156582A (ja) * | 2019-03-25 | 2020-10-01 | キヤノンメディカルシステムズ株式会社 | 超音波診断装置、医用画像処理装置および医用画像処理プログラム |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102013221603A1 (de) * | 2013-10-24 | 2015-04-30 | Siemens Aktiengesellschaft | CT-System mit Recheneinheit und Verfahren zur Rekonstruktion und Befundung von CT-Bilddarstellungen |
JP6778158B2 (ja) | 2017-07-27 | 2020-10-28 | 株式会社日立製作所 | X線ct装置、画像生成方法および画像生成プログラム |
JP6987352B2 (ja) * | 2017-11-17 | 2021-12-22 | 富士フイルムヘルスケア株式会社 | 医用画像処理装置および医用画像処理方法 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001198120A (ja) * | 1999-12-08 | 2001-07-24 | Koninkl Philips Electronics Nv | 再構成画像を組合わせる方法 |
JP3700798B2 (ja) * | 1996-08-19 | 2005-09-28 | 富士写真フイルム株式会社 | 画像処理方法および装置 |
JP2007044275A (ja) * | 2005-08-10 | 2007-02-22 | Hitachi Medical Corp | マルチエナジーx線ct装置 |
JP2010226694A (ja) * | 2009-02-24 | 2010-10-07 | Ricoh Co Ltd | 画像処理装置及び画像処理方法 |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1156451B1 (en) * | 1995-09-29 | 2004-06-02 | Fuji Photo Film Co., Ltd. | Image processing method and apparatus |
US6667815B1 (en) * | 1998-09-30 | 2003-12-23 | Fuji Photo Film Co., Ltd. | Method and apparatus for processing images |
US7409092B2 (en) * | 2002-06-20 | 2008-08-05 | Hrl Laboratories, Llc | Method and apparatus for the surveillance of objects in images |
JP4413504B2 (ja) | 2003-02-13 | 2010-02-10 | 株式会社東芝 | 医用画像処理装置、医用画像処理方法および医用画像処理プログラム |
JP4069943B2 (ja) * | 2003-12-03 | 2008-04-02 | 株式会社ニコン | ノイズ除去の強弱を画面内でコントロールする画像処理装置、画像処理プログラム、画像処理方法、および電子カメラ |
JP4780374B2 (ja) * | 2005-04-21 | 2011-09-28 | Nkワークス株式会社 | 粒状ノイズ抑制のための画像処理方法及びプログラム及びこの方法を実施する粒状抑制処理モジュール |
EP1882449A4 (en) | 2005-05-18 | 2010-05-05 | Hitachi Medical Corp | RADIOGRAPHY DEVICE AND IMAGE PROCESSING PROGRAM |
JP5042465B2 (ja) * | 2005-05-18 | 2012-10-03 | 株式会社日立メディコ | 放射線撮影装置、画像処理方法 |
JP5253835B2 (ja) * | 2008-02-19 | 2013-07-31 | 株式会社キーエンス | 画像生成装置、画像生成方法及びコンピュータプログラム |
KR101481551B1 (ko) * | 2008-06-03 | 2015-01-13 | 엘지전자 주식회사 | 영상 노이즈 제거 장치 및 방법 |
JP5143038B2 (ja) * | 2009-02-02 | 2013-02-13 | オリンパス株式会社 | 画像処理装置及び画像処理方法 |
-
2013
- 2013-03-14 JP JP2014507658A patent/JP6124868B2/ja active Active
- 2013-03-14 IN IN5834DEN2014 patent/IN2014DN05834A/en unknown
- 2013-03-14 CN CN201380005062.5A patent/CN104066378B/zh active Active
- 2013-03-14 WO PCT/JP2013/057118 patent/WO2013146283A1/ja active Application Filing
- 2013-03-14 US US14/372,330 patent/US9454804B2/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3700798B2 (ja) * | 1996-08-19 | 2005-09-28 | 富士写真フイルム株式会社 | 画像処理方法および装置 |
JP2001198120A (ja) * | 1999-12-08 | 2001-07-24 | Koninkl Philips Electronics Nv | 再構成画像を組合わせる方法 |
JP2007044275A (ja) * | 2005-08-10 | 2007-02-22 | Hitachi Medical Corp | マルチエナジーx線ct装置 |
JP2010226694A (ja) * | 2009-02-24 | 2010-10-07 | Ricoh Co Ltd | 画像処理装置及び画像処理方法 |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2015100539A (ja) * | 2013-11-26 | 2015-06-04 | 日立アロカメディカル株式会社 | 超音波診断装置 |
US10217248B2 (en) | 2014-08-27 | 2019-02-26 | General Electric Company | Method for removing streak from detector cell with performance difference |
JP2020078377A (ja) * | 2018-11-12 | 2020-05-28 | 株式会社日立製作所 | 画像再構成装置および画像再構成方法 |
JP7077208B2 (ja) | 2018-11-12 | 2022-05-30 | 富士フイルムヘルスケア株式会社 | 画像再構成装置および画像再構成方法 |
JP2020156582A (ja) * | 2019-03-25 | 2020-10-01 | キヤノンメディカルシステムズ株式会社 | 超音波診断装置、医用画像処理装置および医用画像処理プログラム |
JP7297485B2 (ja) | 2019-03-25 | 2023-06-26 | キヤノンメディカルシステムズ株式会社 | 超音波診断装置、医用画像処理装置および医用画像処理プログラム |
Also Published As
Publication number | Publication date |
---|---|
JPWO2013146283A1 (ja) | 2015-12-10 |
JP6124868B2 (ja) | 2017-05-10 |
US9454804B2 (en) | 2016-09-27 |
CN104066378B (zh) | 2016-10-12 |
US20150010224A1 (en) | 2015-01-08 |
IN2014DN05834A (ja) | 2015-05-15 |
CN104066378A (zh) | 2014-09-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6124868B2 (ja) | 画像処理装置及び画像処理方法 | |
Liu et al. | Total variation-stokes strategy for sparse-view X-ray CT image reconstruction | |
JP7139119B2 (ja) | 医用画像生成装置及び医用画像生成方法 | |
JP5978429B2 (ja) | 医用画像処理装置、医用画像処理方法 | |
US9265475B2 (en) | Methods and apparatus for scatter correction for CBCT system and cone-beam image reconstruction | |
US7379575B2 (en) | Method for post- reconstructive correction of images of a computer tomograph | |
EP3195265B1 (en) | Iterative image reconstruction with a sharpness driven regularization parameter | |
US10395397B2 (en) | Metal artifacts reduction for cone beam CT | |
US8965078B2 (en) | Projection-space denoising with bilateral filtering in computed tomography | |
US8885903B2 (en) | Method and apparatus for statistical iterative reconstruction | |
US10789738B2 (en) | Method and apparatus to reduce artifacts in a computed-tomography (CT) image by iterative reconstruction (IR) using a cost function with a de-emphasis operator | |
JP2017196404A (ja) | 画像処理装置、x線ct装置及び画像処理方法 | |
JP2019516460A (ja) | 空間とスペクトル情報に基づく複数エネルギーのct画像におけるノイズ制御のためのシステムと方法 | |
Zeng et al. | Penalized weighted least-squares approach for multienergy computed tomography image reconstruction via structure tensor total variation regularization | |
JP2019130302A (ja) | 医用画像処理装置及びx線ctシステム | |
WO2014167935A1 (ja) | X線ct装置、再構成演算装置、及び再構成演算方法 | |
Van Aarle et al. | Super-resolution for computed tomography based on discrete tomography | |
US10813616B2 (en) | Variance reduction for monte carlo-based scatter estimation | |
JP2011203160A (ja) | X線ct画像再構成方法及びx線ct画像再構成プログラム | |
CN114387359A (zh) | 一种三维x射线低剂量成像方法及装置 | |
CN112842370A (zh) | 用于x射线成像中的参数噪声调制的方法和系统 | |
CN107106114B (zh) | 运算装置、x射线ct装置及图像重构方法 | |
Xie et al. | Scatter correction for cone-beam computed tomography using self-adaptive scatter kernel superposition | |
JP2013119021A (ja) | X線ct装置及び画像処理方法 | |
US20050018889A1 (en) | Systems and methods for filtering images |
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: 13769416 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014507658 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14372330 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13769416 Country of ref document: EP Kind code of ref document: A1 |