WO2017012528A1 - 一种点扩散函数估计方法及系统 - Google Patents

一种点扩散函数估计方法及系统 Download PDF

Info

Publication number
WO2017012528A1
WO2017012528A1 PCT/CN2016/090336 CN2016090336W WO2017012528A1 WO 2017012528 A1 WO2017012528 A1 WO 2017012528A1 CN 2016090336 W CN2016090336 W CN 2016090336W WO 2017012528 A1 WO2017012528 A1 WO 2017012528A1
Authority
WO
WIPO (PCT)
Prior art keywords
spread function
standard deviation
point spread
directions
rectangular block
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2016/090336
Other languages
English (en)
French (fr)
Inventor
康克军
陈志强
李元景
张丽
赵自然
刘耀红
顾建平
王志明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tsinghua University
Nuctech Co Ltd
Original Assignee
Tsinghua University
Nuctech Co Ltd
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 Tsinghua University, Nuctech Co Ltd filed Critical Tsinghua University
Publication of WO2017012528A1 publication Critical patent/WO2017012528A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/73—Deblurring; Sharpening
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G06T7/41—Analysis of texture based on statistical description of texture
    • G06T7/44—Analysis of texture based on statistical description of texture using image operators, e.g. filters, edge density metrics or local histograms
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/77—Determining position or orientation of objects or cameras using statistical methods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10116—X-ray image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20021—Dividing image into blocks, subimages or windows
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20076—Probabilistic image processing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06T2207/30168—Image quality inspection

Definitions

  • the present invention relates to the field of perspective imaging technology, and in particular, to a point spread function estimation method and system.
  • image blurring due to scanning device motion and detector crosstalk often occurs, which has a certain impact on the visual effects of the image and the examiner's observation of the image content and the discovery of suspicious regions in the image.
  • the image quality degradation of the system can be approximated by a point spread function (PSF) convolution of the image. If the system point spread function is known, the image can be deblurred by some image restoration techniques to obtain a clear image.
  • the quantitative description of the point spread function itself can be used as an important indicator to evaluate the imaging quality of the system. Through its in-depth analysis, it can also provide guidance for the improvement of the system software and hardware.
  • the invention provides a point spread function estimation method and system, which solves the technical problem that the point spread function estimation method in the prior art is not high in automation and slow in execution speed.
  • the present invention provides a point spread function estimation method, which includes include:
  • Scanning imaging is performed on a plurality of rectangular blocks of the same size and closely arranged by a scanning device having different materials and/or different thicknesses, wherein an incident angle of the rays is perpendicular to the scanning direction during scanning imaging, and The incident angle of the ray is perpendicular to the closely arranged surfaces of the plurality of rectangular blocks;
  • the standard deviation parameters of the two directions are combined to obtain parameters of the two-dimensional point spread function to estimate the point spread function.
  • the side length of the rectangular block is not less than a predetermined multiple of the detector size of the scanning device.
  • the acquiring the line spread function along the length and the width of the rectangular block according to the scanned image further includes:
  • the standard deviation parameters respectively obtained by the line diffusion function include:
  • Gaussian fitting of the line spread functions in the two directions yields a standard deviation parameter of the line spread function.
  • the method further includes:
  • the standard deviation parameters of the plurality of line spread functions in the same direction are averaged, and the average standard deviation parameter of the direction is obtained by combining the material and the mass thickness of the rectangular block.
  • the present invention also provides a point spread function estimation system, comprising:
  • An imaging unit configured to scan and image a plurality of rectangular blocks of the same size and closely arranged by using a scanning device, wherein the plurality of rectangular blocks are made of different materials and/or different thicknesses, wherein the angle of incidence and scanning of the rays during scanning imaging The direction is vertical, and an incident angle of the ray is perpendicular to a closely aligned surface of the plurality of rectangular blocks;
  • a parameter obtaining unit configured to acquire a line spread function along two directions of length and width of the rectangular block according to the scanned image, and obtain standard deviation parameters of the line spread function respectively;
  • An estimating unit is configured to combine the standard deviation parameters of the two directions to obtain a parameter of the two-dimensional point spread function to estimate the point spread function.
  • the side length of the rectangular block is not less than a predetermined multiple of the detector size of the scanning device.
  • the parameter obtaining unit is further configured to:
  • the parameter obtaining unit is further configured to:
  • Gaussian fitting of the line spread functions in the two directions yields a standard deviation parameter of the line spread function.
  • system further includes:
  • An averaging unit connected to the parameter obtaining unit and the estimating unit, for averaging standard deviation parameters of a plurality of line spreading functions in the same direction, and combining the material and mass thickness of the rectangular block to obtain the direction Average standard deviation parameter.
  • the present invention provides an image processing method for an X-ray imaging apparatus, comprising the steps of the point spread function estimating method according to any of the above.
  • the present invention also provides an image processing system for an X-ray imaging apparatus that employs an image processing method for an X-ray imaging apparatus as described above.
  • the present invention provides an image processing system for an X-ray imaging apparatus, comprising a radiation source, a detector, and a rectangular block, and the point spread function estimating system according to any one of the preceding claims.
  • the material of the rectangular block is one or more of graphite, lead, iron, and polyethylene.
  • the rectangular block has a mass thickness of 2 to 50 g/cm 2 .
  • the length and/or width of the rectangular block is such that the detector effectively collects X-rays 5-100 times the size.
  • the point spread function estimation method and system provided by the present invention, it is possible to estimate the point spread function of the scanning device without knowing the thickness and size of the rectangular block mass and without knowing the size of the detector, and it is possible to provide a A point spread function estimation method for automatic and fast scanning devices.
  • the invention utilizes the quantitative description of the estimated system point spread function to give a quantitative evaluation of the image imaging quality of the scanning device, and deblurs the image by image restoration technology, thereby improving the image clarity, and Corresponding suggestions are made for the hardware improvement of imaging of the scanning device according to the parameters of the point spread function.
  • the point spread function estimation of the present invention applicable to various scanning imaging systems is particularly suitable for the estimation of a point spread function which can approximate a Gaussian shape but different bidirectional standard deviation parameters.
  • FIG. 1 is a schematic diagram showing the basic flow of a point spread function estimation method according to an embodiment of the present invention
  • FIG. 2 is a schematic flow chart of a preferred embodiment of a method for estimating a point spread function according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of a rectangular block scan image according to Embodiment 1 of the present invention.
  • step 202 in Embodiment 1 of the present invention is a schematic diagram of specific execution steps of step 202 in Embodiment 1 of the present invention.
  • Figure 5 is a schematic view showing a calculation area of a transition curve of cumulative gray scale change in the vertical direction in Embodiment 1 of the present invention.
  • FIG. 6 is a schematic diagram showing a calculation area of a transition curve of cumulative gray scale change in the horizontal direction in Embodiment 1 of the present invention.
  • FIG. 8 is a schematic diagram of a line spread function curve obtained by differentially dividing a cumulative gray-scale transition curve shown in FIG. 7 in Embodiment 1 of the present invention.
  • FIG. 10 is a schematic flow chart of a preferred embodiment of a point spread function estimating method according to Embodiment 2 of the present invention.
  • FIG. 11 is a schematic diagram showing the basic structure of a point spread function estimating system in Embodiment 3 of the present invention.
  • Figure 12 is a block diagram showing the configuration of an image processing system for an X-ray imaging apparatus in Embodiment 6 of the present invention.
  • the embodiment of the present invention first provides a point spread function estimation method. Referring to FIG. 1, the method includes:
  • Step 101 Scanning and imaging a plurality of rectangular blocks of the same size and closely arranged by using a scanning device having different materials and/or different thicknesses, wherein the incident angle of the rays is perpendicular to the scanning direction during scanning imaging. And an incident angle of the ray is perpendicular to a closely aligned surface of the plurality of rectangular blocks;
  • Step 102 Obtain a line spread function along two directions of length and width of the rectangular block according to the scanned image, and obtain standard deviation parameters of the line spread function respectively.
  • Step 103 Combine the standard deviation parameters of the two directions to obtain parameters of the two-dimensional point spread function to estimate the point spread function.
  • the side length of the rectangular block is not less than the detector size of the scanning device.
  • the predetermined multiple such as the side length of the rectangular block is 5-100 times the size of the detector of the scanning device.
  • the detector size refers to the size of the detector to effectively collect X-rays.
  • the obtaining a line spread function (LSF) along the length and the width of the rectangular block according to the scan image may further include: determining each rectangle according to the number and arrangement manner of the rectangular block The position of the boundary between the blocks, the cumulative gray-scale change transition curve is calculated for each of the two-direction boundary, and the cumulative gray-scale change transition curve is differentiated to obtain the line spread function of the two directions.
  • LSF line spread function
  • respectively obtaining the standard deviation parameter of the line spread function may include: performing a Gaussian fitting on the line spread functions of the two directions to obtain a standard deviation parameter of the line spread function.
  • the method may further include: a criterion for multiple line diffusion functions in the same direction
  • the difference parameter is averaged and combined with the material and mass thickness of the rectangular block to obtain an average standard deviation parameter for this direction.
  • Embodiment 1 of the present invention provides a method for estimating a point spread function by using a plurality of rectangular blocks of substantially uniform plate-like materials having substantially the same size and different thicknesses.
  • the method includes:
  • Step 201 Scanning imaging is performed on a plurality of rectangular blocks of the same size and closely arranged by a scanning device.
  • each material block has a size of 200 mm ⁇ 200 mm, and the first row is an iron material having a mass thickness of 6 g/cm 2 , 8 g/cm 2 , and 10 g/cm 2 respectively; the first row of the second row is a mass thickness of 4 g/cm.
  • the iron material of 2 , the second and third are respectively polyethylene materials having a mass thickness of 2 g/cm 2 and 4 g/cm 2 ; the first row of the third row is an iron material having a mass thickness of 3 g/cm 2 , and the second Three polyethylene materials each having a mass thickness of 6 g/cm 2 and 8 g/cm 2 were used.
  • This closely-arranged material was placed in a scanning device whose surface was perpendicular to the incident direction of the rays, and scanned to obtain a scanned image in which the scanning direction was parallel to the horizontal side direction of the rectangular block in Fig. 3.
  • Step 202 Acquire a line spread function along two directions of length and width of the rectangular block according to the scanned image, and obtain standard deviation parameters of the line spread function respectively.
  • step 402 a rectangular region centered on the midpoint of the boundary is selected near the boundary of each rectangular block to calculate a cumulative grayscale transition curve. If the calculation area of the vertical boundary between the first block and the second block in the upper row shown in FIG. 5 is selected, the calculation area of the horizontal boundary between the second block in the middle row and the second block in the lower row shown in FIG. 6 is subsequently Accumulating along the boundary direction results in a cumulative gray-scale transition curve corresponding to the vertical and horizontal directions, respectively.
  • FIG. 7 shows a cumulative gradation change transition curve obtained by longitudinally accumulating the calculation area shown in FIG. 5. This curve can calculate the first-order difference to obtain the approximate derivative, and change the gray value of the two ends of the transition curve according to the gray level to determine whether to use the forward difference or the backward difference (when the left gray value is small, the direction is adopted).
  • the front difference, ie f(x)-f(x-1); otherwise, the backward difference, ie f(x)-f(x+1)) is used to ensure that most of the calculated difference values are Positive, thus obtaining an approximate line spread function curve, as shown in FIG.
  • Step 203 Combine the standard deviation parameters of the two directions to obtain parameters of the two-dimensional point spread function to estimate the point spread function.
  • the distribution of the point spread function PSF can be estimated as:
  • C is the normalization coefficient
  • ⁇ x and ⁇ y are the standard deviation parameters of the x direction and the y direction, respectively.
  • the method for estimating the point spread function in Embodiment 2 of the present invention is substantially the same as that in Embodiment 1, wherein before the step 1004 (corresponding to Step 203 of Embodiment 1 of the present invention), the method further includes:
  • Step 1003 Average the standard deviation parameters of the plurality of line spread functions in the same direction, and combine the material and the mass thickness of the rectangular block to obtain an average standard deviation parameter of the direction.
  • the standard deviation of the two directions can be respectively performed.
  • the parameters are averaged to obtain the mean standard deviation parameters for the horizontal and vertical directions.
  • step 1004 the standard deviation parameters ⁇ x , ⁇ y of the x-direction and y-direction used to estimate the point spread function are also replaced by the mean standard deviation parameters in the x-direction and the y-direction, respectively.
  • Embodiment 3 of the present invention provides a point spread function estimation system. Referring to FIG. 11, the method includes:
  • the imaging unit 1101 is configured to perform scanning imaging on a plurality of rectangular blocks of the same size and closely arranged by using a scanning device, wherein the plurality of rectangular blocks are made of different materials and/or different thicknesses, wherein the angle of incidence of the rays during scanning imaging is The scanning direction is perpendicular, and an incident angle of the ray is perpendicular to a closely arranged surface of the plurality of rectangular blocks;
  • a parameter obtaining unit 1102 configured to acquire a line spread function along two directions of length and width of the rectangular block according to the scanned image, and obtain standard deviation parameters of the line spread function respectively;
  • the estimating unit 1103 is configured to combine the standard deviation parameters of the two directions to obtain parameters of the two-dimensional point spread function to estimate the point spread function.
  • the side length of the rectangular block is not less than a predetermined multiple of the detector size of the scanning device.
  • the parameter obtaining unit 1102 is further configured to: determine a position of a boundary between each rectangular block according to the number and arrangement manner of the rectangular block, and calculate a cumulative gray transition curve for the boundary of the two directions, and The cumulative gray-scale change transition curve is differentiated to obtain a line spread function in the two directions.
  • the parameter obtaining unit 1102 is further configured to: perform a Gaussian fitting on the line diffusion functions in the two directions to obtain a standard deviation parameter of the line diffusion function.
  • the system may further include: an averaging unit respectively connected to the parameter obtaining unit and the estimating unit, configured to average standard deviation parameters of a plurality of line spreading functions in the same direction, and combine the rectangular blocks The material and mass thickness are obtained to obtain the mean standard deviation parameter for this direction.
  • an averaging unit respectively connected to the parameter obtaining unit and the estimating unit, configured to average standard deviation parameters of a plurality of line spreading functions in the same direction, and combine the rectangular blocks The material and mass thickness are obtained to obtain the mean standard deviation parameter for this direction.
  • Embodiment 4 of the present invention is directed to an image processing method for an X-ray imaging apparatus that applies a point spread function estimation method according to an embodiment of the present invention to determine an image quality of an X-ray imaging apparatus or perform image restoration.
  • the pixel range of the point spread function may be small, such as 3 ⁇ 3 pixels; and when the image quality is poor, the pixel range of the point spread function may be correspondingly larger, such as 7 ⁇ 7 pixels.
  • the obtained point spread function can be used to perform corresponding operations on the acquired image to recover a better quality image and improve image quality.
  • image restoration such as deconvolution, which can obtain a better image by deconvolving the point spread function from the acquired image.
  • Embodiment 5 of the present invention provides an image processing system for an X-ray imaging apparatus using an image processing method for an X-ray imaging apparatus in Embodiment 4 of the present invention.
  • Embodiment 6 of the present invention provides an image processing system for an X-ray imaging apparatus.
  • a radiation source 1201, a detector 1202, a rectangular block 1203, and a point spread function estimating system 1204 according to Embodiment 3 of the present invention are provided. .
  • the radiation source 1201 emits X-rays, illuminates a plurality of closely arranged rectangular blocks 1203, collects signals through the detector 1202, and analyzes the collected signals through a point spread function estimation system 1204 for point spread function estimation. .
  • the material of the rectangular block may be one or more of graphite, lead, iron, and polyethylene.
  • the rectangular block may have a mass thickness ranging from 2 to 50 g/cm 2 .
  • the length and/or width of the rectangular block may be 5-100 times the effective collection of X-ray dimensions by the detector.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Image Processing (AREA)
  • Measurement Of Radiation (AREA)

Abstract

本发明提供一种点扩散函数估计方法和系统，所述方法包括：对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像，所述多个矩形块的制作材料不同和/或质量厚度不同，其中扫描成像时射线的入射角度与扫描方向垂直，且所述射线的入射角度垂直于所述多个矩形块的紧密排列的表面；根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数；将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。本发明自动化程度高，执行速度快。

Description

一种点扩散函数估计方法及系统 技术领域
本发明涉及透视成像技术领域，尤其涉及一种点扩散函数估计方法及系统。
背景技术
在透视成像系统中，经常会出现由于扫描装置运动和探测器串扰引起的图像模糊情况，对图像视觉效果和检查人员观察图像内容及发现图像中的可疑区域造成一定的影响。系统的图像质量退化可以近似为一个点扩散函数(PSF，Point Spreading Function)对图像的卷积，如果系统点扩散函数已知，可以通过一些图像复原技术对图像进行去模糊处理，从而得到清晰的图像。另外，点扩散函数的定量描述本身可以作为评价系统成像质量的一个重要指标，通过对它的深入分析还可以为系统软件硬件的改进提供指导意见。
为了改善图像质量和分析图像退化原因，需要估计系统的点扩散函数。但在以往的估计方法中，大多数是针对可见光成像系统的，如典型倒谱法、变分贝叶斯法、稀疏约束最优化方法等。这些方法都针对一般图像，没有利用特定装置，估计精度较差，且大多数方法计算速度非常慢。并且针对X-射线成像系统的方法往往需要精确已知测试装置和系统的一些参数，比如缝的宽度、孔的半径、探测器的物理尺寸等等。如果测试装置设计不合理或加工精度不高，容易造成估计不准确，计算过程也比较复杂。
发明内容
本发明提供一种点扩散函数估计方法及系统，以解决现有技术中点扩散函数估计方法自动化程度不高、执行速度慢的技术问题。
为解决上述技术问题，本发明提供一种点扩散函数估计方法，包 括：
对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像，所述多个矩形块的制作材料不同和/或质量厚度不同，其中扫描成像时射线的入射角度与扫描方向垂直，且所述射线的入射角度垂直于所述多个矩形块的紧密排列的表面；
根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数；
将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。
进一步地，所述矩形块的边长不小于所述扫描装置的探测器尺寸的预定倍数。
进一步地，所述根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数还包括：
根据所述矩形块的个数和排布方式确定各个矩形块之间边界的位置，对两个方向的边界分别计算累积灰度变化过渡曲线，并对所述累积灰度变化过渡曲线进行差分得到所述两个方向的线扩散函数。
进一步地，所述分别得到所述线扩散函数的标准差参数包括：
对所述两个方向的线扩散函数进行高斯拟合别得到所述线扩散函数的标准差参数。
进一步地，在所述分别得到所述线扩散函数的标准差参数之后，所述将两个方向的所述标准差参数结合之前，还包括：
对同一方向的多个线扩散函数的标准差参数进行平均，并结合所述矩形块的材料和质量厚度，获取此方向的平均标准差参数。
另一方面，本发明还提供一种点扩散函数估计系统，包括：
成像单元，用于对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像，所述多个矩形块的制作材料不同和/或质量厚度不同，其中扫描成像时射线的入射角度与扫描方向垂直，且所述射线的入射角度垂直于所述多个矩形块的紧密排列的表面；
参数获取单元，用于根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数；
估计单元，用于将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。
进一步地，所述矩形块的边长不小于所述扫描装置的探测器尺寸的预定倍数。
进一步地，所述参数获取单元还用于：
根据所述矩形块的个数和排布方式确定各个矩形块之间边界的位置，对两个方向的边界分别计算累积灰度变化过渡曲线，并对所述累积灰度变化过渡曲线进行差分得到所述两个方向的线扩散函数。
进一步地，所述参数获取单元还用于：
对所述两个方向的线扩散函数进行高斯拟合别得到所述线扩散函数的标准差参数。
进一步地，所述系统还包括：
平均单元，与所述参数获取单元和所述估计单元分别相连，用于对同一方向的多个线扩散函数的标准差参数进行平均，并结合所述矩形块的材料和质量厚度，获取此方向的平均标准差参数。
再一方面，本发明还提供一种用于X射线成像装置的图像处理方法，包括如上任一项所述的点扩散函数估计方法的步骤。
另一方面，本发明还提供一种用于X射线成像装置的图像处理系统，其采用如上所述的用于X射线成像装置的图像处理方法。
再一方面，本发明还提供一种用于X射线成像装置的图像处理系统，包括辐射源、探测器和矩形块以及如权利要求上任一项所述的点扩散函数估计系统。
进一步地，所述矩形块的材料为石墨、铅、铁、聚乙烯中的一种或多种。
进一步地，所述矩形块的质量厚度为2-50g/cm2。
进一步地，所述矩形块的长和/或宽为所述探测器有效收集X射线 尺寸的5-100倍。
可见，在本发明所提供的点扩散函数估计方法和系统中，能够在无需已知矩形块质量厚度和大小，也无需已知探测器尺寸的情况下估计扫描装置的点扩散函数，可以提供一种自动、快速的扫描装置的点扩散函数估计方式。本发明利用估计得到的系统点扩散函数的定量描述，可以对扫描装置的图像成像质量给出一个定量的评价，通过图像复原技术以对图像进行去模糊处理，从而提高图像的清晰度，并可根据点扩散函数的参数对扫描装置成像的硬件改进提出相应的建议。本发明适用于各种扫描成像系统的点扩散函数估计尤其适用于可近似为高斯形状、但双向标准差参数不同的点扩散函数的估计。
附图说明
为了更清楚地说明本发明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一简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例点扩散函数估计方法的基本流程示意图；
图2是本发明实施例1点扩散函数估计方法的一个优选实施例流程示意图；
图3是本发明实施例1的矩形块扫描图像示意图；
图4是本发明实施例1中步骤202的具体执行步骤示意图；
图5是本发明实施例1中竖直方向累积灰度变化过渡曲线的计算区域示意图；
图6是本发明实施例1中水平方向累积灰度变化过渡曲线的计算区域示意图；
图7是本发明实施例1中图5所示区域经计算得到的累积灰度变化过渡曲线；
图8是本发明实施例1中图7所示累积灰度变化过渡曲线经差分得到的线扩散函数曲线示意图；
图9是本发明实施例1中图8所示的线扩散函数曲线经去除噪声干扰后的曲线(实线)及高斯拟合结果(虚线)；
图10是本发明实施例2的点扩散函数估计方法的一个优选实施例流程示意图；
图11是本发明实施例3中点扩散函数估计系统的基本结构示意图；
图12是本发明实施例6中用于X射线成像装置的图像处理系统的结构示意图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明实施例首先提供一种点扩散函数估计方法，参见图1，包括：
步骤101：对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像，所述多个矩形块的制作材料不同和/或质量厚度不同，其中扫描成像时射线的入射角度与扫描方向垂直，且所述射线的入射角度垂直于所述多个矩形块的紧密排列的表面；
步骤102：根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数；
步骤103：将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。
可选地，所述矩形块的边长不小于所述扫描装置的探测器尺寸的 预定倍数，如矩形块的边长为扫描装置的探测器尺寸的5-100倍。其中，探测器尺寸指探测器有效收集X射线的尺寸。
可选地，根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数(LSF，Line Spreading Function)还可以包括：根据所述矩形块的个数和排布方式确定各个矩形块之间边界的位置，对两个方向的边界分别计算累积灰度变化过渡曲线，并对所述累积灰度变化过渡曲线进行差分得到所述两个方向的线扩散函数。
可选地，分别得到所述线扩散函数的标准差参数可以包括：对所述两个方向的线扩散函数进行高斯拟合别得到所述线扩散函数的标准差参数。
可选地，在所述分别得到所述线扩散函数的标准差参数之后，所述将两个方向的所述标准差参数结合之前，还可以包括：对同一方向的多个线扩散函数的标准差参数进行平均，并结合所述矩形块的材料和质量厚度，获取此方向的平均标准差参数。
实施例1：
本发明实施例1提供了利用多块大小基本相同、质量厚度不同的均匀板状材料的矩形块进行点扩散函数估计的方法，具体参见图2，包括：
步骤201：对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像。
本步骤中，利用由不同质量厚度的铁和聚乙烯材料块组成的多个矩形块进行扫描成像，扫描图像如图3所示。其中各材料块大小均为200mm×200mm，第1排为质量厚度分别为6g/cm2、8g/cm2、10g/cm2的铁材料；第2排第1个为质量厚度为4g/cm2的铁材料、第2、3个分别为质量厚度为2g/cm2和4g/cm2的聚乙烯材料；第3排第1个为质量厚度为3g/cm2的铁材料、第2、3个分别为质量厚度为6g/cm2和8g/cm2的聚乙烯材料。将这一经紧密排列固定的材料放置在表面垂直于射线入射方向的扫描装置中进行扫描，获得扫描图像，其中扫描 方向与图3中矩形块的水平边方向平行。
步骤202：根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数。
本步骤的具体执行流程参见图4。其中，根据已知的上述多个矩形块每行及每列块个数信息，可在步骤401中通过简单计算得到每个矩形块边界的位置。进而，在步骤402中，在每个矩形块边界附近选择以边界中点为中心的一个矩形区域用以计算累积灰度变化过渡曲线。如选择图5所示的上排第一块和第二块之间竖直边界的计算区域，图6所示的中排第二块和下排第二块之间水平边界的计算区域，随后沿边界走向累加得到竖直和水平两个方向分别对应的累积灰度变化过渡曲线。并在步骤403中得到两个方向的线扩散函数。例如，图7表示图5所示的计算区域经纵向累加得到的累积灰度变化过渡曲线。对此曲线可以计算一阶差分从而得到近似导数，并根据灰度变化过渡曲线两端的灰度值的大小，以确定采用向前差分还是向后差分(其中当左边灰度值较小时，采用向前差分，即f(x)-f(x-1)；反之，则采用向后差分，即f(x)-f(x+1))，以保证经计算得到的绝大多数差分值为正，从而得到近似线扩散函数曲线，如图8所示。
在步骤403获取两个方向的线扩散函数之后，分别从其最大值向两边搜索，遇到第一个负值或零值后，将其后所有值置零，以得到去除噪声干扰的线扩散函数曲线。其中图8中的线扩散函数曲线去除噪声干扰之后如图9中实线所示，对图9中的实线在步骤404中以高斯形状拟合，计算其均值和标准差，获取线扩散函数的标准差参数，拟合结果如图9中虚线所示。
步骤203：将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。
本步骤中，将水平和竖直(即x和y)两个方向的标准差参数进行结合，可以估计点扩散函数PSF的分布为：
Figure PCTCN2016090336-appb-000001
其中C为归一化系数，σx、σy分别为x方向和y方向的标准差参数。
实施例2：
参见图10，本发明实施例2中的点扩散函数估计的方法与实施例1中基本相同，其中，在步骤1004(对应本发明实施例1步骤203)之前，还包括：
步骤1003：对同一方向的多个线扩散函数的标准差参数进行平均，并结合所述矩形块的材料和质量厚度，获取此方向的平均标准差参数。
本步骤中，对本发明实施例1中所有6个内部竖直块边界和6个内部水平块边界进行前述步骤的线扩散函数计算和标准差参数拟合后，可以分别对两个方向的标准差参数进行平均，得到水平和竖直方向的平均标准差参数。
与此对应地，在步骤1004中，估计点扩散函数所采用的x方向和y方向的标准差参数σx、σy也将相应替换为x方向和y方向的平均标准差参数。
实施例3：
本发明实施例3提供了一种点扩散函数估计系统，参见图11，包括：
成像单元1101，用于对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像，所述多个矩形块的制作材料不同和/或质量厚度不同，其中扫描成像时射线的入射角度与扫描方向垂直，且所述射线的入射角度垂直于所述多个矩形块的紧密排列的表面；
参数获取单元1102，用于根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数；
估计单元1103，用于将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。
其中，所述矩形块的边长不小于所述扫描装置的探测器尺寸的预定倍数。
其中，参数获取单元1102还可以用于：根据所述矩形块的个数和排布方式确定各个矩形块之间边界的位置，对两个方向的边界分别计算累积灰度变化过渡曲线，并对所述累积灰度变化过渡曲线进行差分得到所述两个方向的线扩散函数。
其中，参数获取单元1102还可以用于：对所述两个方向的线扩散函数进行高斯拟合别得到所述线扩散函数的标准差参数。
其中，所述系统还可以包括：平均单元，与所述参数获取单元和所述估计单元分别相连，用于对同一方向的多个线扩散函数的标准差参数进行平均，并结合所述矩形块的材料和质量厚度，获取此方向的平均标准差参数。
实施例4：
本发明实施例4通过一种用于X射线成像装置的图像处理方法，其应用了如本发明实施例所述的点扩散函数估计方法，以判断X射线成像装置的成像质量或进行图像复原。
其中，当得到所估计的点扩散函数之后，可以通过预设的像素范围阈值判断成像是否清晰，成像质量是否满足要求。当所成图像清晰时，点扩散函数的像素范围可能较小，如为3×3像素；而当成像质量较差时，点扩散函数的像素范围可能相应较大，如为7×7像素。
另外，当所获取的图像质量不尽如人意时，可以利用所得到的点扩散函数与所获取的图像进行相应运算，以复原得到质量较好的图像，提高成像质量。图像复原的运算方法可以有多种，例如可以为反卷积运算，其通过将点扩散函数与所获取的图像进行反卷积，可以得到较为理想的图像。
实施例5：
本发明实施例5提供一种利用本发明实施例4中的用于X射线成像装置的图像处理方法的用于X射线成像装置的图像处理系统。
实施例6：
本发明实施例6提供一种用于X射线成像装置的图像处理系统，参见图12，包括辐射源1201、探测器1202、矩形块1203以及如本发明实施例3中的点扩散函数估计系统1204。
其中，辐射源1201发射X射线，照射到多个紧密排列的矩形块1203上，并通过探测器1202收集信号，将所收集的信号通过点扩散函数估计系统1204进行分析，以进行点扩散函数估计。
可选地，矩形块的材料可以为石墨、铅、铁、聚乙烯中的一种或多种。
可选地，矩形块的质量厚度范围可以为2-50g/cm2。
可选地，矩形块的长和/或宽可以为探测器有效收集X射线尺寸的5-100倍。
可见，在本发明实施例所提供的点扩散函数估计方法和系统中，至少具有如下有益效果：
(1)使用简单、自动化程度高。仅要求各个材料块为厚度均匀、大小相同的矩形块，但对具体厚度和大小并没有精确要求，降低了加工要求；除扫描图像外，无需人工干预，自动完成区域定位和参数计算等过程；
(2)鲁棒性好、自适应能力强。综合利用多个区域间的阶跃边缘，受噪声影响较小；仅要求各个材料块为厚度均匀、大小基本相同的矩形块，并已知两个方向材料块的数目。但对具体厚度和大小并没有精确要求，降低了对材料厚度、尺寸等物理加工精度的要求；
(3)执行速度快。由于点扩散函数近似为高斯形状，估计过程中不需大量反卷积操作，处理速度得到了明显的提高。
最后应说明的是：以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领 域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims (16)

  1. 一种点扩散函数估计方法，其特征在于，包括：
    对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像，所述多个矩形块的制作材料不同和/或质量厚度不同，其中扫描成像时射线的入射角度与扫描方向垂直，且所述射线的入射角度垂直于所述多个矩形块的紧密排列的表面；
    根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数；
    将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。
  2. 根据权利要求1所述的点扩散函数估计方法，其特征在于：
    所述矩形块的边长不小于所述扫描装置的探测器尺寸的预定倍数。
  3. 根据权利要求1所述的点扩散函数估计方法，其特征在于，所述根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数还包括：
    根据所述矩形块的个数和排布方式确定各个矩形块之间边界的位置，对两个方向的边界分别计算累积灰度变化过渡曲线，并对所述累积灰度变化过渡曲线进行差分得到所述两个方向的线扩散函数。
  4. 根据权利要求1所述的点扩散函数估计方法，其特征在于，所述分别得到所述线扩散函数的标准差参数包括：
    对所述两个方向的线扩散函数进行高斯拟合别得到所述线扩散函数的标准差参数。
  5. 根据权利要求1所述的点扩散函数估计方法，其特征在于，在所述分别得到所述线扩散函数的标准差参数之后，所述将两个方向的所述标准差参数结合之前，还包括：
    对同一方向的多个线扩散函数的标准差参数进行平均，并结合所 述矩形块的材料和质量厚度，获取此方向的平均标准差参数。
  6. 一种点扩散函数估计系统，其特征在于，包括：
    成像单元，用于对大小相同且紧密排列的多个矩形块利用扫描装置进行扫描成像，所述多个矩形块的制作材料不同和/或质量厚度不同，其中扫描成像时射线的入射角度与扫描方向垂直，且所述射线的入射角度垂直于所述多个矩形块的紧密排列的表面；
    参数获取单元，用于根据扫描图像获取沿所述矩形块的长和宽两个方向的线扩散函数，并分别得到所述线扩散函数的标准差参数；
    估计单元，用于将两个方向的标准差参数结合，得到二维点扩散函数的参数，以估计所述点扩散函数。
  7. 根据权利要求6所述的点扩散函数估计系统，其特征在于：
    所述矩形块的边长不小于所述扫描装置的探测器尺寸的预定倍数。
  8. 根据权利要求6所述的点扩散函数估计系统，其特征在于，所述参数获取单元还用于：
    根据所述矩形块的个数和排布方式确定各个矩形块之间边界的位置，对两个方向的边界分别计算累积灰度变化过渡曲线，并对所述累积灰度变化过渡曲线进行差分得到所述两个方向的线扩散函数。
  9. 根据权利要求6所述的点扩散函数估计系统，其特征在于，所述参数获取单元还用于：
    对所述两个方向的线扩散函数进行高斯拟合别得到所述线扩散函数的标准差参数。
  10. 根据权利要求6所述的点扩散函数估计系统，其特征在于，所述系统还包括：
    平均单元，与所述参数获取单元和所述估计单元分别相连，用于对同一方向的多个线扩散函数的标准差参数进行平均，并结合所述矩形块的材料和质量厚度，获取此方向的平均标准差参数。
  11. 一种用于X射线成像装置的图像处理方法，其特征在于，包括如权利要求1-5中任一项所述的点扩散函数估计方法的步骤。
  12. 一种用于X射线成像装置的图像处理系统，其特征在于，采用如权利要求11所述的用于X射线成像装置的图像处理方法。
  13. 一种用于X射线成像装置的图像处理系统，其特征在于，包括辐射源、探测器和矩形块以及如权利要求6-10中任一项所述的点扩散函数估计系统。
  14. 如权利要求13所述的用于X射线成像装置的图像处理系统，其特征在于：所述矩形块的材料为石墨、铅、铁、聚乙烯中的一种或多种。
  15. 如权利要求13所述的用于X射线成像装置的图像处理系统，其特征在于：所述矩形块的质量厚度为2-50g/cm2。
  16. 如权利要求13所述的用于X射线成像装置的图像处理系统，其特征在于：所述矩形块的长和/或宽为所述探测器有效收集X射线尺寸的5-100倍。
PCT/CN2016/090336 2015-07-23 2016-07-18 一种点扩散函数估计方法及系统 Ceased WO2017012528A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510437211.6 2015-07-23
CN201510437211.6A CN106372035B (zh) 2015-07-23 2015-07-23 一种点扩散函数估计方法及系统

Publications (1)

Publication Number Publication Date
WO2017012528A1 true WO2017012528A1 (zh) 2017-01-26

Family

ID=57833710

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/090336 Ceased WO2017012528A1 (zh) 2015-07-23 2016-07-18 一种点扩散函数估计方法及系统

Country Status (3)

Country Link
US (1) US10152774B2 (zh)
CN (1) CN106372035B (zh)
WO (1) WO2017012528A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106372035B (zh) * 2015-07-23 2018-12-18 清华大学 一种点扩散函数估计方法及系统
EP3680851B1 (en) * 2019-01-09 2021-11-03 Agfa Nv Method and system to calculate the point spread function of a digital image detector system based on a mtf modulated quantum-noise measurement
CN111488701B (zh) * 2020-06-28 2020-10-16 中国人民解放军国防科技大学 一种基于点扩散函数的计算机辅助装调方法、装置及设备
CN117274103B (zh) * 2023-10-11 2025-12-02 南京工程学院 基于光斑图像估计点扩散函数的离焦图像复原方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101291391A (zh) * 2007-04-20 2008-10-22 致伸科技股份有限公司 图像处理方法以及相关的部分点扩散函数估测方法
CN101354307A (zh) * 2008-09-22 2009-01-28 哈尔滨工业大学 动态目标调制传递函数测量方法与装置
CN102819830A (zh) * 2012-08-15 2012-12-12 北京交通大学 一种新的基于卡尔曼滤波的点扩散函数估计方法
CN103927740A (zh) * 2014-02-18 2014-07-16 北京象开科技有限公司 获得图像的点扩展函数的方法和装置

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7251306B2 (en) * 2004-11-17 2007-07-31 General Electric Company Methods, apparatus, and software to facilitate iterative reconstruction of images
US8249357B2 (en) * 2006-10-23 2012-08-21 Ben Gurion University Of The Negev, Research And Development Authority Blind restoration of images degraded by isotropic blur
US8326011B2 (en) * 2008-05-21 2012-12-04 Varian Medical Systems, Inc. Methods, systems, and computer-program products for estimating scattered radiation in radiographic projections
WO2010104969A1 (en) * 2009-03-11 2010-09-16 Zoran Corporation Estimation of point spread functions from motion-blurred images
KR102020531B1 (ko) * 2012-08-20 2019-09-10 삼성전자주식회사 선형 감마선원을 이용하여 고해상도의 ｐｅｔ(양전자 방출 단층 촬영) 영상을 생성하는 방법 및 장치
US9785860B2 (en) * 2014-07-16 2017-10-10 The Cleveland Clinic Foundation Real-time image enhancement for X-ray imagers
US9779491B2 (en) * 2014-08-15 2017-10-03 Nikon Corporation Algorithm and device for image processing
CN106372035B (zh) * 2015-07-23 2018-12-18 清华大学 一种点扩散函数估计方法及系统
KR101850871B1 (ko) * 2015-08-26 2018-04-23 주식회사 디알텍 방사선 영상의 처리방법 및 방사선 촬영시스템
KR20170088681A (ko) * 2016-01-25 2017-08-02 삼성전자주식회사 단층 촬영 장치 및 그에 따른 단층 영상 복원 방법

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101291391A (zh) * 2007-04-20 2008-10-22 致伸科技股份有限公司 图像处理方法以及相关的部分点扩散函数估测方法
CN101354307A (zh) * 2008-09-22 2009-01-28 哈尔滨工业大学 动态目标调制传递函数测量方法与装置
CN102819830A (zh) * 2012-08-15 2012-12-12 北京交通大学 一种新的基于卡尔曼滤波的点扩散函数估计方法
CN103927740A (zh) * 2014-02-18 2014-07-16 北京象开科技有限公司 获得图像的点扩展函数的方法和装置

Also Published As

Publication number Publication date
US20170024862A1 (en) 2017-01-26
CN106372035A (zh) 2017-02-01
CN106372035B (zh) 2018-12-18
US10152774B2 (en) 2018-12-11

Similar Documents

Publication Publication Date Title
JP4601134B2 (ja) 形状特徴に基づく欠陥検出の方法及び装置
US11672471B2 (en) Osteoporosis diagnostic support apparatus
US8761540B2 (en) Method and system for estimating noise level
CN114373086B (zh) 集成模板匹配方法、装置、计算机设备及存储介质
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
CN117764983A (zh) 一种用于智能制造产线的双目识别的视觉检测的方法
WO2017012528A1 (zh) 一种点扩散函数估计方法及系统
CN114972084A (zh) 一种图像对焦准确度评价方法及系统
US7751643B2 (en) Method and apparatus for removing uneven brightness in an image
CN106447684A (zh) 工业ct图像中弱边缘尺寸测量方法
WO2015133123A1 (ja) 放射線画像処理装置および方法並びにプログラム
US20210192251A1 (en) Method and system for selecting image region that facilitates blur kernel estimation
CN105844594A (zh) 一种工业ct扫描系统旋转中心校正方法
US7508994B2 (en) Method for detecting streaks in digital images
CN106651768B (zh) 图像校正方法及装置、x射线摄影设备
CN105261048A (zh) 一种小球中心锥束投影位置的精确定位方法
CN109813259B (zh) 高动态x射线成像方法、存储介质和装置
CN115661020B (zh) 一种dr胸腔图像肋骨检测的方法及装置
Gea et al. Region of interest micro-CT of the middle ear: A practical approach
JP2017070590A (ja) 画像処理装置及びその制御方法、コンピュータプログラム
CN114298934B (zh) 一种基于像素调节的面颊夹显影弱化方法、装置
CN118552638A (zh) 一种工业ct散射校正方法
JP2017176384A (ja) 放射線撮影装置、放射線撮影システム、放射線撮影方法、及びプログラム
Peskin et al. Estimating the Number of Manually Segmented Cellular Objects Required to Evaluate the Performance of a Segmentation Algorithm
CN1983331A (zh) 校正计算机断层造影中因过度曝光引起的图像伪影的方法

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: 16827219

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: 16827219

Country of ref document: EP

Kind code of ref document: A1