CN116739914A - Deconvolution method and system for image geometry sharpness correction - Google Patents
Deconvolution method and system for image geometry sharpness correction Download PDFInfo
- Publication number
- CN116739914A CN116739914A CN202310493965.8A CN202310493965A CN116739914A CN 116739914 A CN116739914 A CN 116739914A CN 202310493965 A CN202310493965 A CN 202310493965A CN 116739914 A CN116739914 A CN 116739914A
- Authority
- CN
- China
- Prior art keywords
- image
- sharpness
- radiological
- spread function
- point spread
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000012937 correction Methods 0.000 title claims description 7
- 230000005855 radiation Effects 0.000 claims abstract description 46
- 230000006870 function Effects 0.000 claims description 32
- 238000012545 processing Methods 0.000 claims description 18
- 238000001914 filtration Methods 0.000 claims description 17
- 238000001514 detection method Methods 0.000 claims description 16
- 235000002566 Capsicum Nutrition 0.000 claims description 14
- 239000006002 Pepper Substances 0.000 claims description 14
- 150000003839 salts Chemical class 0.000 claims description 14
- 235000016761 Piper aduncum Nutrition 0.000 claims description 13
- 235000017804 Piper guineense Nutrition 0.000 claims description 13
- 235000008184 Piper nigrum Nutrition 0.000 claims description 13
- 238000005315 distribution function Methods 0.000 claims description 10
- 244000203593 Piper nigrum Species 0.000 claims 1
- 238000009659 non-destructive testing Methods 0.000 abstract description 8
- 238000004458 analytical method Methods 0.000 abstract description 6
- 241000722363 Piper Species 0.000 description 12
- 238000006243 chemical reaction Methods 0.000 description 11
- 230000002285 radioactive effect Effects 0.000 description 10
- 241000519995 Stachys sylvatica Species 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 9
- 239000000463 material Substances 0.000 description 9
- 239000002245 particle Substances 0.000 description 9
- 238000003384 imaging method Methods 0.000 description 8
- 230000005251 gamma ray Effects 0.000 description 6
- 238000007689 inspection Methods 0.000 description 6
- 239000000126 substance Substances 0.000 description 5
- 230000015556 catabolic process Effects 0.000 description 4
- 238000006731 degradation reaction Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000035515 penetration Effects 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 241000758706 Piperaceae Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000006249 magnetic particle Substances 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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
- G06T2207/10116—X-ray image
-
- 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
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Measurement Of Radiation (AREA)
Abstract
The invention provides a deconvolution method and a deconvolution system for correcting geometric sharpness of an image, comprising the following steps: calculating geometrical sharpness of the radiological image based on the relative positions of the radiological source, the sample and the image detector and the exit parameters of the radiological source; estimating a point spread function expressed by a two-dimensional Cauchy distribution model according to the geometric acuteness; and restoring the radiation image by adopting an RL algorithm according to the point spread function. The deconvolution method and the deconvolution system for correcting the geometric sharpness of the image can solve the problem of geometric sharpness of the existing neutron image, and achieve the purpose of improving the nondestructive testing analysis precision.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a deconvolution method and a deconvolution system for correcting geometrical sharpness of an image.
Background
The neutron photography technique plays an important role in the image information acquisition link of nondestructive detection analysis. In a neutron imaging system, since the ideal point neutron source is not present, an inherent blur will be generated at the foreground edge of the neutron image after the sample is imaged, and this blur is called geometric sharpness of the neutron image, resulting in a degradation of the imaging quality of the neutron image. Meanwhile, the neutron and the detected sample undergo nuclear reaction to release gamma rays, and a gamma ray white spot background is superimposed on an image detector, so that further degradation of a neutron image is caused.
Disclosure of Invention
In view of the above problems, the present invention has been made to provide a deconvolution method and system for correcting geometric sharpness of an image, which overcomes the above problems or at least partially solves the above problems, and can solve the geometric sharpness problem of the existing neutron image, thereby achieving the purpose of improving the accuracy of nondestructive testing analysis.
Specifically, the invention provides a deconvolution method for correcting geometric sharpness of an image, which comprises the following steps:
calculating geometrical sharpness of the radiological image based on the relative positions of the radiological source, the sample and the image detector and the exit parameters of the radiological source;
estimating a point spread function expressed by a two-dimensional Cauchy distribution model according to the geometric acuteness;
and restoring the radiation image by adopting an RL algorithm according to the point spread function.
Optionally, the deconvolution method further includes:
detecting a plaque area from the radiation image by using an edge algorithm, wherein the plaque area comprises white plaque areas polluted by gamma rays and/or noise plaques polluted by salt and pepper noise;
and carrying out median filtering processing on the radiological image to remove the plaque area in the radiological image, so as to obtain a speckle-removed radiological image for restoration.
Optionally, said calculating geometric sharpness of the radiological image includes:
obtaining and noting the relative distance between the radiation source and the image detector as a first distance, the relative distance between the sample and the image detector as a second distance, and the exit diameter of the radiation source;
and according to the first distance, the second distance and the outlet diameter, obtaining the geometric sharpness by applying a similar triangle theorem.
Optionally, the estimating a point spread function expressed by using a two-dimensional cauchy distribution model includes:
taking the geometric non-sharpness as the full width at half maximum of the two-dimensional cauchy distribution function, and calculating the dispersity parameter of the point spread function;
and obtaining a point spread function expressed by adopting a two-dimensional Cauchy distribution model according to the dispersity parameter.
Optionally, the calculating the dispersity parameter of the point spread function includes:
obtaining a length discrete value, a width discrete value and a discrete value of each point spread function;
and normalizing the length discrete value, the width discrete value and the discrete values to obtain the dispersity parameter.
Optionally, the detecting plaque area from the radiological image using an edge algorithm includes:
detecting the radiation image by using a Laplacian operator to obtain detection values of all positions of the radiation image;
and taking the corresponding area with the detection value larger than a set threshold value as the plaque area.
Optionally, the median filtering processing is performed on the plaque area and is introduced into an RL algorithm, so that an RA-RL algorithm for image restoration is obtained, wherein the expression is as follows:
wherein: i t Representing the result of the t-th iteration, k represents the point spread function, k T Indicating the inversion of k in the horizontal direction, and L (·) indicating the laplace operator detection noise point.
Optionally, the median filtering processing is performed on the radiological image, including:
differentiating a useful region outside the plaque region on the radiological image;
and carrying out the median filtering processing on the plaque area and carrying out the retaining processing on the useful area.
Optionally, the radiological image is a neutron image or an X-Ray image.
The invention also provides a deconvolution system for image geometry sharpness correction, comprising a memory, a processor and a machine executable program stored on the memory and running on the processor, and the processor implementing the deconvolution method according to any of the preceding claims when executing the machine executable program.
In the deconvolution method for correcting the geometric sharpness of the image, firstly, the geometric sharpness of the radioactive image is calculated, a point spread function expressed by a two-dimensional Cauchy distribution model is estimated according to the geometric sharpness, and finally, the radioactive image is restored by adopting an RL algorithm according to the point spread function, so that the geometric sharpness problem of the existing radioactive image can be solved, and the aim of improving the nondestructive testing analysis precision is fulfilled.
The above, as well as additional objectives, advantages, and features of the present invention will become apparent to those skilled in the art from the following detailed description of a specific embodiment of the present invention when read in conjunction with the accompanying drawings.
Drawings
Some specific embodiments of the invention will be described in detail hereinafter by way of example and not by way of limitation with reference to the accompanying drawings. The same reference numbers will be used throughout the drawings to refer to the same or like parts or portions. It will be appreciated by those skilled in the art that the drawings are not necessarily drawn to scale. In the accompanying drawings:
FIG. 1 is a schematic flow chart of a deconvolution method according to one embodiment of the invention;
FIG. 2 is a schematic flow chart of step S100 according to one embodiment of the invention;
FIG. 3 is a schematic flow chart of an image detection system according to one embodiment of the invention;
FIG. 4 is a schematic flow chart of step S200 according to one embodiment of the invention;
FIG. 5 is a schematic flow chart diagram of a deconvolution method in accordance with one embodiment of the present invention;
FIG. 6 is a schematic flow chart of step S400 according to one embodiment of the invention;
FIG. 7 is a schematic flow chart of step S500 according to one embodiment of the invention;
FIG. 8 is a general flow chart of RA-RL algorithm processing of radiological images according to one embodiment of the present invention.
Detailed Description
Deconvolution methods and systems for image geometry sharpness correction in accordance with embodiments of the present invention are described below with reference to fig. 1-8. In the description of the present embodiment, it should be understood that the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature, i.e. one or more such features. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise. When a feature "comprises or includes" a feature or some of its coverage, this indicates that other features are not excluded and may further include other features, unless expressly stated otherwise.
Unless specifically stated or limited otherwise, the terms "disposed," "mounted," "connected," "secured," "coupled," and the like should be construed broadly, as they may be connected, either permanently or removably, or integrally; can be mechanically or electrically connected; either directly or indirectly, through intermediaries, or both, may be in communication with each other or in interaction with each other, unless expressly defined otherwise. Those of ordinary skill in the art will understand the specific meaning of the terms described above in the present invention as the case may be.
Furthermore, in the description of the present embodiments, a first feature "above" or "below" a second feature may include the first and second features being in direct contact, or may include the first and second features not being in direct contact but being in contact through another feature therebetween. That is, in the description of the present embodiment, the first feature being "above", "over" and "upper" the second feature includes the first feature being directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature "under", "beneath", or "under" a second feature may be a first feature directly under or diagonally under the second feature, or simply indicate that the first feature is less level than the second feature.
In the description of the present embodiment, a description referring to the terms "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Nondestructive testing is also called nondestructive testing, and is a technology for detecting defects, chemical and physical parameters of materials, parts and equipment by adopting principle technologies such as rays, ultrasound, infrared and electromagnetism and combining instruments on the premise of not damaging or affecting the service performance of a detected object. The conventional nondestructive inspection methods are five types of eddy current inspection (ECT), radiographic inspection (RT), ultrasonic inspection (UT), magnetic particle inspection (MT) and liquid penetration inspection (PT). The common radiation used in radiographic detection is X-Ray, which is a particle stream generated by the transition of electrons in atoms between two energy levels with greatly different energies, and is electromagnetic radiation with wavelengths between ultraviolet rays and gamma rays. The wavelength of which is very short, between about 0.01 and 100 angstroms. The X-Ray penetration is extremely strong, and the difference of Ray absorption can be caused by the density, the material and the like of the object. After passing through the object rapidly, the uniform X-Ray will form an unevenly distributed image, which is a projection of the internal structure of the object, also called X-Ray image. By utilizing such characteristics, the inside of the sample can be inspected without damaging the sample.
In recent years, neutron photography technology is widely applied in the field of nondestructive detection and achieves good effects, and the neutron photography technology is widely applied in product quality control in the departments of aerospace, aviation, chemical industry metallurgy, nuclear industry and the like, scientific research and bioscience, and shows unique effects.
Neutron photography is based on the fundamental principle that radiation attenuates as it passes through an object. When a neutron is incident on a sample to be irradiated, the intensity and spatial distribution of the transmitted neutron will change due to interactions, such as scattering and nuclear reactions, of the neutron with the nuclei in the sample.
Because different materials have different attenuation characteristics on neutron beams, the strength of the action is related to the properties (such as density holes of constituent elements) of the materials contained in the sample in the local area where the action occurs, so that the transmitted neutron beams contain information of the internal components and structures of the sample, and then the spatial distribution of the transmitted neutron fluence rate is displayed by using a specific technology and a related imaging technology to obtain a neutron image, thus obtaining the comprehensive information of various defects of the spatial distribution density change of the materials contained in the sample, which is the basic principle of the neutron photography technology. Neutron photography uses interaction imaging of neutrons and atomic nuclei, and has unique advantages not possessed by other nondestructive testing methods in terms of penetrability, resolution of elements with similar atomic numbers, and the like.
In X-Ray or neutron photographic systems, since the ideal point radiation source is not present, an inherent blur, known as geometric sharpness of the radiological image, will be created at the foreground edge of the radiological image after imaging the sample, causing a reduction in the imaging quality of the radiological image. Meanwhile, the particles generated by the radioactive source and the detected sample undergo nuclear reaction to release gamma rays, and a gamma ray white spot background is superimposed on an image detector, so that the radiographic image is further degraded.
In view of the above problems, the present invention has been made to provide a deconvolution method and system for correcting geometric sharpness of an image, which overcomes the above problems or at least partially solves the above problems, and can solve the geometric sharpness problem of the existing neutron image, thereby achieving the purpose of improving the accuracy of nondestructive testing analysis.
Fig. 1 is a schematic flow chart of a deconvolution method according to an embodiment of the present invention, as shown in fig. 1, and referring to fig. 2 to 8, the embodiment of the present invention provides a deconvolution method for image geometry sharpness correction, comprising:
step S100: calculating geometric sharpness of the radiological image based on the relative positions of the radiological source, the sample, and the image detector, and the exit parameters of the radiological source;
step S200: estimating a point spread function expressed by a two-dimensional Cauchy distribution model according to the geometric sharpness;
step S300: and restoring the radiological image by using an RL algorithm according to the point spread function.
Wherein the radiation source is a device capable of generating free particles. For example, when the radiation source is a neutron source, the neutron source may produce free neutrons. The neutron sources may include reactor neutron sources, accelerator neutron sources, neutron tube neutron sources, and isotope neutron sources. Since neutrons are uncharged and cannot be focused, a collimated neutron beam is typically obtained by collimating neutrons from a neutron source with a collimator in order to convert the neutrons into a beam of rays that can be used for neutron imaging.
Neutrons can be classified into cold neutrons, thermal neutrons, resonant neutrons, fast neutrons, etc. according to the energy. The thermal neutron photography has high resolution, is most commonly applied to neutron photography at present, and is mature in technology. However, thermal neutrons can only penetrate thin metal layers, and it is difficult to obtain a clear image for objects such as shells. Fast neutrons can penetrate thicker metal layers, so fast neutron photography can make up for the deficiencies of thermal neutrons in this regard. However, the resolution of fast neutron photography is inferior to that of thermal neutrons, and fast neutron photography has been greatly developed and widely used in recent years because of its unique advantages and practical needs.
Image detectors are key components for detecting the spatial distribution of particles transmitted through a sample. For example, when the radiation source is a neutron source, the image detector is a neutron detector, also known as a neutron conversion screen in neutron photography. Because neutrons cannot be directly detected by a detector such as a film, and the neutron conversion screen contains neutron conversion substances and fluorescent substances, the neutron conversion substances can absorb thermal neutrons or fast neutrons and emit alpha, beta or gamma rays or recoil protons after interaction. These secondary rays or charged particles cause the fluorescent material to emit light, thereby creating a detectable image of the material. The neutron conversion screen comprises a thermal neutron conversion screen, a fast neutron conversion screen and the like.
In step S100, the geometric sharpness of the radiological image can be calculated by the relative positions of the radiation source, the sample and the image detector, and the exit parameters of the radiation source, so that the calculation is relatively simple.
In step S200, the point spread function (Point Spread Function, PSF) is the light field distribution of the output image of the optical system when the input object is a point light source. The functional model for representing the PSF includes Gaussian distribution, kexil distribution, and the like. According to the invention, a two-dimensional Cauchy distribution model is adopted, and the reason is that in the quantum world, particles are far away from each other, for example, the distance from an electron to an atomic nucleus is far away from the size of the particles, and the two-dimensional Cauchy distribution model is adopted, so that the value range is wider, and therefore, the two-dimensional Cauchy distribution model can remarkably describe the position distribution of the particles.
In step S300, the RL algorithm is a non-blind deconvolution image restoration method proposed by Richardson and Lucy. The RL algorithm assumes that the image follows poisson distribution, and adopts a maximum likelihood method for estimation, so that the RL algorithm is an iterative algorithm. The expression for the RL algorithm is:
wherein: i degraded For radiological images, I t Representing the result of the t-th iteration, k represents the point spread function PSF, k T Indicating the inversion of k in the horizontal direction, which is a convolution operation, it can be seen from the equation that as t increases continuously, I t+1 The probability converges to I and the shell recovers the radiological image.
In the deconvolution method for correcting the geometric sharpness of the image, the geometric sharpness of the radiation image is calculated firstly, the point spread function expressed by the two-dimensional Cauchy distribution model is estimated according to the geometric sharpness, and finally the radiation image is restored by adopting the RL algorithm according to the point spread function, so that the problem of the geometric sharpness of the existing radiation image can be solved, and the aim of improving the nondestructive testing analysis precision is fulfilled.
In some embodiments of the present invention, as shown in fig. 2, calculating geometric sharpness of the radiological image in step S100 includes:
step S110: obtaining a relative distance between the radiation source 100 and the image detector 300 and noting the first distance, a relative distance between the sample 200 and the image detector 300 and noting the second distance, and an exit diameter of the radiation source 100;
step S120: and according to the first distance, the second distance and the outlet diameter, using a similar triangle theorem to obtain geometric sharpness.
As shown in fig. 3, the radiation source 100, the sample 200, and the image detector 300 are sequentially arranged in the same direction, wherein the relative distance between the radiation source 100 and the image detector 300 is a first distance and is set to L, the relative distance between the sample 200 and the image detector 300 is a second distance and is set to D, and the exit diameter of the radiation source 100 is set to D.
The ideal source 100 is a point source 100, but in practice the source 100 will be sized so that the ideal source 100 does not exist. As shown in fig. 3, after imaging the sample 200, an inherent blur will be generated at the foreground edge of the radiological image, which blur is referred to as the geometrical sharpness of the radiological image, which is designated as G.
Specifically, as can be seen from fig. 3, the line between two ends of the line segment where the exit diameter D of the radiation source 100 is located and a point of the edge of the sample 200 forms a first triangle, two sides of the first triangle are respectively extended from the point and two intersection points are formed with the image detector 300, and the length of the line segment between the two intersection points is the geometric sharpness G. The line segment between the two extension lines and the two intersection points forms a second triangle. Obviously, the first triangle and the second triangle are similar triangles, and can be obtained by the similar triangle theorem
Can be arranged to obtainGeometry sharpness
Since the first distance L, the second distance D and the exit diameter D of the radiation source 100 are both constant, geometric sharpness G can be obtained.
In some embodiments of the present invention, as shown in fig. 4, estimating a point spread function expressed using a two-dimensional cauchy distribution model in step S200 includes:
step S210: taking the geometric sharpness as the full width at half maximum of the two-dimensional cauchy distribution function, and calculating the dispersity parameter of the point spread function;
step S220: and obtaining a point spread function expressed by adopting a two-dimensional Cauchy distribution model according to the dispersity parameter.
In step S210, the expression of the two-dimensional cauchy distribution function is:
wherein: cauchy (x, y) represents a two-dimensional Cauchy distribution function and ζ represents a dispersity parameter.
The full width at half maximum, FWHM, refers to the distance between the intersection of a straight line passing through the midpoint of the peak height of a functional image and the two sides of the peak, and being parallel to the bottom of the peak. The full width at half maximum of the two-dimensional cauchy distribution function refers to the distance between the line passing through the midpoint of the peak height of the two-dimensional cauchy distribution function and intersecting the two sides of the peak of the two-dimensional cauchy distribution function and making a line parallel to the bottom of the peak. Taking the geometric sharpness G calculated in the embodiment as the full width at half maximum of the two-dimensional cauchy distribution function, the relationship between the geometric sharpness G and the dispersity parameter xi of the point spread function can be calculated as follows: g 2 ＝4(2 2/3 -1)ξ 2 。
In step S220, ζ is substituted into the expression of the two-dimensional cauchy distribution function, so as to obtain the point spread function PSF expressed by using the two-dimensional cauchy distribution model.
In some embodiments of the present invention, the calculating of the dispersion parameter of the point spread function in step S210 includes:
step S211, obtaining a length discrete value, a width discrete value and a discrete value of each point diffusion function;
step S212, normalizing the length discrete value, the width discrete value and the discrete values in each place to obtain a dispersity parameter.
The normalization process refers to scaling the data to fall within a small specific interval, for example, uniformly mapping the data into the [0,1] interval. After normalization processing, the speed of solving the optimal solution can be increased, and the calculation accuracy can be improved.
In some embodiments of the present invention, as shown in fig. 5, the deconvolution method further includes:
step S400: detecting a plaque area from the radiation image by using an edge algorithm, wherein the plaque area comprises white plaque areas polluted by gamma rays and/or noise plaques polluted by salt and pepper noise;
step S500: and carrying out median filtering treatment on the radioactive image to remove plaque areas in the radioactive image, so as to obtain a speckle-removed radioactive image for restoration.
In step S400, the white spot area is formed because gamma rays are released due to the nuclear reaction between the radioactive particles and the sample 200, and the white spot background of gamma rays is superimposed in the radioactive image, resulting in degradation of the radioactive image.
Salt-and-pepper noise (also known as impulse noise) randomly changes some pixel values, which appears on a binary image as white and black pixels. Salt and pepper noise is white and black alternate bright and dark spot noise generated by an image sensor, a transmission channel, decoding processing and the like. Salt and pepper noise is often caused by image cutting. Salt and pepper noise refers to two types of noise, one is salt noise (salt noise) and the other is pepper noise (peppers noise). The salt is white, which means high gray noise, and the pepper is black, which means low gray noise. Generally, two kinds of noise occur simultaneously, and the noise appears on the radiological image as black and white miscellaneous points. Salt and pepper noise can further cause degradation of the radiological image.
In step S500, the median filtering method is a nonlinear smoothing technique, which sets the gray value of each pixel in the radiological image to be the median of the gray values of all pixels in a certain neighborhood window of the point. The median filtering is a nonlinear signal processing technology capable of effectively suppressing noise based on a sequencing statistical theory, and the basic principle of median filtering is to replace the value of a point in a digital image or a digital sequence with the median of the point values in a neighborhood of the point, so that surrounding pixel values are close to a true value, and isolated noise points are eliminated. The plaque area in the radiation image can be effectively removed by adopting a median filtering method, and the speckle-removed radiation image for restoration is obtained.
In some embodiments of the present invention, as shown in fig. 6, the detecting plaque area from the radiological image using the edge algorithm in step S400 includes:
step S410: detecting the radiation image by using a Laplacian operator to obtain detection values of all positions of the radiation image;
step S420: and taking the corresponding area with the detection value larger than the set threshold value as the plaque area.
The expression for detecting the radiation image by using the Laplacian operator is as follows: l (I) degraded ) L (-) represents the detection of noise points by the Laplace operator, so as to obtain detection values of all positions of the radiation image, and the plaque area is found out by comparing the detection values of all positions with a set threshold value. Specifically, a corresponding region where the detection value is larger than the set threshold value is taken as a plaque region. For example, the plaque region is defined as γ (i×k), and the non-plaque region is defined as
In some embodiments of the present invention, median filtering processing is performed on plaque areas and introduced into an RL algorithm, so as to obtain an RA-RL algorithm for image restoration, where the expression is as follows:
in this embodiment, the RA-RL algorithm may also be referred to as an adaptive RL algorithm, which removes gamma ray white spots and/or salt and pepper noise prior to deconvolution of the radiological image, and suppresses ringing effects due to gamma ray white spots. The RA-RL algorithm not only can correct geometric sharpness of the radiation image, but also can remove white spots and/or salt and pepper noise of gamma rays, thereby ensuring the overall quality of the radiation image.
In some embodiments of the present invention, an overall flow chart of the RA-RL algorithm for the radiation image processing is shown in fig. 8, where the Original image (Original image) is the radiation image in the above embodiment, and the Restored image (Restored image) is the image after the restoration of the radiation image.
In some embodiments of the present invention, as shown in fig. 7, the median filtering processing is performed on the radiological image in step S500, including:
step S510: differentiating a useful region located outside the plaque region on the radiological image;
step S520: and carrying out median filtering treatment on the plaque area and carrying out retention treatment on the useful area.
The useful region in step S510 is the non-plaque region in the above embodiment. And carrying out median filtering treatment on the plaque area to remove gamma ray white spots and/or salt and pepper noise in the radiation image, and not carrying out treatment on the useful area to obtain the radiation image with the gamma ray white spots and/or salt and pepper noise removed.
The embodiment of the invention also provides a deconvolution system for correcting geometric sharpness of an image, which comprises a memory, a processor and a machine executable program stored on the memory and running on the processor, wherein the deconvolution method according to any embodiment is realized when the processor executes the machine executable program.
By now it should be appreciated by those skilled in the art that while a number of exemplary embodiments of the invention have been shown and described herein in detail, many other variations or modifications of the invention consistent with the principles of the invention may be directly ascertained or inferred from the present disclosure without departing from the spirit and scope of the invention. Accordingly, the scope of the present invention should be understood and deemed to cover all such other variations or modifications.
Claims (10)
1. A deconvolution method for image geometry sharpness correction, comprising:
calculating geometrical sharpness of the radiological image based on the relative positions of the radiological source, the sample and the image detector and the exit parameters of the radiological source;
estimating a point spread function expressed by a two-dimensional Cauchy distribution model according to the geometric acuteness;
and restoring the radiation image by adopting an RL algorithm according to the point spread function.
2. The deconvolution method of claim 1, further comprising:
detecting a plaque area from the radiation image by using an edge algorithm, wherein the plaque area comprises white plaque areas polluted by gamma rays and/or noise plaques polluted by salt and pepper noise;
and carrying out median filtering processing on the radiological image to remove the plaque area in the radiological image, so as to obtain a speckle-removed radiological image for restoration.
3. The deconvolution method of claim 1, wherein,
the calculating of geometric sharpness of the radiological image includes:
obtaining and noting the relative distance between the radiation source and the image detector as a first distance, the relative distance between the sample and the image detector as a second distance, and the exit diameter of the radiation source;
and according to the first distance, the second distance and the outlet diameter, obtaining the geometric sharpness by applying a similar triangle theorem.
4. The deconvolution method of claim 1, wherein,
the estimating the point spread function expressed by adopting a two-dimensional cauchy distribution model comprises the following steps:
taking the geometric non-sharpness as the full width at half maximum of the two-dimensional cauchy distribution function, and calculating the dispersity parameter of the point spread function;
and obtaining a point spread function expressed by adopting a two-dimensional Cauchy distribution model according to the dispersity parameter.
5. The deconvolution method of claim 4, wherein,
the calculating the dispersity parameter of the point spread function comprises the following steps:
obtaining a length discrete value, a width discrete value and a discrete value of each point spread function;
and normalizing the length discrete value, the width discrete value and the discrete values to obtain the dispersity parameter.
6. The deconvolution method of claim 2, wherein,
the detecting plaque area from the radiological image by using an edge algorithm includes:
detecting the radiation image by using a Laplacian operator to obtain detection values of all positions of the radiation image;
and taking the corresponding area with the detection value larger than a set threshold value as the plaque area.
7. The deconvolution method of claim 6, wherein,
introducing the median filtering processing to the plaque area into an RL algorithm to obtain an RA-RL algorithm for image restoration, wherein the expression is as follows:
wherein: i t Representing the result of the t-th iteration, k tableShows the point spread function, k T Indicating the inversion of k in the horizontal direction, and L (·) indicating the laplace operator detection noise point.
8. The deconvolution method of claim 2, wherein,
the median filtering processing of the radiological image comprises the following steps:
differentiating a useful region outside the plaque region on the radiological image;
and carrying out the median filtering processing on the plaque area and carrying out the retaining processing on the useful area.
9. The deconvolution method of claim 1, wherein,
the radiation image is a neutron image or an X-Ray image.
10. A deconvolution system for image geometry sharpness correction, characterized in that it comprises a memory, a processor and a machine executable program stored on the memory and running on the processor, and in that the processor implements the deconvolution method according to any of claims 1 to 9 when executing the machine executable program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310493965.8A CN116739914A (en) | 2023-04-28 | 2023-04-28 | Deconvolution method and system for image geometry sharpness correction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310493965.8A CN116739914A (en) | 2023-04-28 | 2023-04-28 | Deconvolution method and system for image geometry sharpness correction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116739914A true CN116739914A (en) | 2023-09-12 |
Family
ID=87903334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310493965.8A Pending CN116739914A (en) | 2023-04-28 | 2023-04-28 | Deconvolution method and system for image geometry sharpness correction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116739914A (en) |
-
2023
- 2023-04-28 CN CN202310493965.8A patent/CN116739914A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6879715B2 (en) | Iterative X-ray scatter correction method and apparatus | |
| CN100457039C (en) | X-ray scatter correction | |
| Yu et al. | Finite detector based projection model for high spatial resolution | |
| JP3566762B2 (en) | Apparatus and method for computed tomography | |
| US8009892B2 (en) | X-ray image processing system | |
| Yang et al. | Automatic X-ray inspection for escaped coated particles in spherical fuel elements of high temperature gas-cooled reactor | |
| US7916828B1 (en) | Method for image construction | |
| Yan et al. | Comparison of heuristic and deterministic algorithms in neutron coded imaging reconstruction | |
| CN116739914A (en) | Deconvolution method and system for image geometry sharpness correction | |
| Coban et al. | Assessing the efficacy of tomographic reconstruction methods through physical quantification techniques | |
| Ali et al. | Image reconstruction techniques using projection data from transmission method | |
| US10078150B2 (en) | Detecting and quantifying materials in containers utilizing an inverse algorithm with adaptive regularization | |
| Movafeghi et al. | Using nonlocal operators for measuring dimensions of defects in radiograph of welded objects | |
| Piault et al. | A thresholding based iterative reconstruction method for limited-angle tomography data | |
| He et al. | IDART: An improved discrete tomography algorithm for reconstructing images with multiple gray levels | |
| Movafeghi et al. | Using interpolation with nonlocal autoregressive modeling for defect detection in welded objects | |
| Konovalov et al. | Spatial resolution of few-view computed tomography using algebraic reconstruction techniques | |
| Trampert et al. | Spherically symmetric volume elements as basis functions for image reconstructions in computed laminography | |
| Gopi et al. | Iterative computed tomography reconstruction from sparse-view data | |
| Li et al. | A MLEM-TV-MRP algorithm for fast neutron computed tomography reconstruction of high statistical noise and sparse sampling | |
| Abir | Iterative CT reconstruction from few projections for the nondestructive post irradiation examination of nuclear fuel assemblies | |
| Zhou et al. | Investigation of maximum a posteriori probability expectation-maximization for image-based weighting spectral X-ray CT image reconstruction | |
| Van Rompu et al. | Some terrestial tardigrades from Zimbabwe | |
| Zhang et al. | A high-density gamma white spots-Gaussian mixture noise removal method for neutron images denoising based on Swin Transformer UNet and Monte Carlo calculation | |
| Fu | Performance enhancement approaches for a dual energy X-ray imaging system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |