CN114758066B - Image processing method, apparatus, device, storage medium, and computer program product - Google Patents
Image processing method, apparatus, device, storage medium, and computer program product Download PDFInfo
- Publication number
- CN114758066B CN114758066B CN202210322761.3A CN202210322761A CN114758066B CN 114758066 B CN114758066 B CN 114758066B CN 202210322761 A CN202210322761 A CN 202210322761A CN 114758066 B CN114758066 B CN 114758066B
- Authority
- CN
- China
- Prior art keywords
- target object
- data
- image
- target
- projection images
- 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.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 30
- 238000004590 computer program Methods 0.000 title claims abstract description 20
- 239000000463 material Substances 0.000 claims abstract description 59
- 238000004088 simulation Methods 0.000 claims abstract description 58
- 238000012545 processing Methods 0.000 claims abstract description 56
- 238000001228 spectrum Methods 0.000 claims abstract description 34
- 238000000034 method Methods 0.000 claims abstract description 30
- 238000003384 imaging method Methods 0.000 claims abstract description 20
- 238000005070 sampling Methods 0.000 claims description 49
- 230000005855 radiation Effects 0.000 claims description 28
- 238000012937 correction Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 229910000838 Al alloy Inorganic materials 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 241000282414 Homo sapiens Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 239000000956 alloy Substances 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image processing method, an image processing device, a storage medium and a computer program product, wherein the method comprises the following steps: reading the modeling file to obtain point-plane data of the target object; constructing simulation body data of the target object according to the material and point-plane data of the target object and the energy spectrum curve of a ray source radiating the target object; performing ray projection simulation on simulation body data of a target object to obtain N initial projection images; simulating scattering and noise in the actual imaging process, and carrying out fuzzy processing on N Zhang Chushi projection images to obtain N intermediate projection images; and superposing the target projection image on the target background image according to the lambert law, and carrying out attenuation treatment on the superposed image so as to output a final image obtained by the attenuation treatment as a data set picture, wherein the target projection image is at least one of N intermediate projection images. Therefore, the problems that a great deal of time and economic cost are consumed in the process of manufacturing the data set picture in the prior art can be solved.
Description
Technical Field
The present invention relates to the field of image processing, and in particular, to an image processing method, apparatus, device, computer storage medium, and computer program product.
Background
In order to improve the speed of image examination and the accuracy of identification of a target object (e.g., contraband), image identification methods based on deep learning have been widely used in the field of image detection using a radiation source (e.g., X-rays). When image recognition is performed based on a deep learning technology, an image dataset is required to be used, the number and the quality of the image dataset are critical to the accuracy of image recognition, but are influenced by design parameters of an imaging system where a ray source is located, and the image quality difference generated by different imaging systems is extremely large, so that a great amount of time and economic cost are required to manufacture a dataset picture.
Disclosure of Invention
The embodiment of the invention provides an image processing method, an image processing device, image processing equipment, a computer storage medium and a computer program product, which can solve the problem that a great deal of time and economic cost are required to be consumed in the process of manufacturing a data set picture in the prior art.
In a first aspect, there is provided an image processing method, the method comprising:
reading a modeling file to obtain point-plane data of a target object, wherein the modeling file is obtained by modeling the target object;
constructing simulation body data of the target object according to the material of the target object, the point-surface data and an energy spectrum curve of a ray source radiating the target object;
Performing ray projection simulation on the simulation body data of the target object to obtain N initial projection images, wherein N is a positive integer, and the N initial projection images correspond to different target object arrangement positions;
simulating scattering and noise in the actual imaging process, and carrying out fuzzy processing on N initial projection images to obtain N intermediate projection images;
And superposing the target projection image on the target background image according to the lambert law, and carrying out attenuation processing on the superposed image so as to output a final image obtained by the attenuation processing as a data set picture, wherein the target projection image is at least one of N intermediate projection images.
Optionally, the constructing the simulation body data of the target object according to the material of the target object, the point-plane data and the energy spectrum curve of the ray source irradiating the target object includes:
Performing space sampling and connected domain filling processing on the point-plane data of the target object to obtain binary data of the target object;
And carrying out material assignment on the binary volume data through the energy spectrum curve of the radiation source radiating the target object and the material of the target object to obtain the simulation volume data of the target object.
Optionally, the point-plane data includes M triangular patches, where M is a positive integer; the step of performing spatial sampling and connected domain filling processing on the point-plane data of the target object to obtain binary data of the target object comprises the following steps:
Performing space discrete sampling on the point-plane data of the target object to obtain a plurality of sampling points;
traversing a plurality of sampling points, and assigning the volume data of the sampling points on the triangular patch as a first numerical value;
And judging the three-dimensional connected domain of the sampling points which are not positioned on the triangular surface patch, so as to assign the volume data of the sampling points which accord with the three-dimensional connected domain to the first value, and assign the volume data of the sampling points which do not accord with the three-dimensional connected domain to the second value, wherein the first value is not equal to the second value.
Optionally, the performing material assignment on the binary data through the energy spectrum curve of the radiation source radiating the target object and the material of the target object to obtain the simulation body data of the target object includes:
According to the energy spectrum curve of the ray source, determining the linear attenuation coefficients of different materials in the target object when the ray source irradiates the target object;
and carrying out material assignment on binary volume data of target object areas corresponding to different materials according to the line attenuation coefficients of the different materials to obtain simulation volume data of the target object.
Optionally, the simulating the scattering and noise in the actual imaging process performs fuzzy processing on the N initial projection images to obtain N intermediate projection images, including:
respectively carrying out convolution blurring processing on N initial projection images;
And adding noise to each initial projection image after convolution blurring processing to obtain N intermediate projection images.
Optionally, before the superimposing the target projection image on the background image, the method further includes:
Obtaining M typical background images;
Carrying out zero-load correction on M typical background images to obtain M background images; wherein the M background images include the target background image.
In a second aspect, there is provided an image processing apparatus comprising:
the reading module is used for reading a modeling file to obtain point-plane data of a target object, wherein the modeling file is obtained by modeling the target object;
The construction module is used for constructing simulation body data of the target object according to the material of the target object, the point-surface data and the energy spectrum curve of a ray source radiating the target object;
The simulation module is used for carrying out ray projection simulation on the simulation body data of the target object to obtain N initial projection images, wherein N is a positive integer, and the N initial projection images correspond to different target object arrangement positions;
the fuzzy processing module is used for simulating scattering and noise in the actual imaging process, and performing fuzzy processing on the N initial projection images to obtain N intermediate projection images;
And the superposition attenuation module is used for superposing the target projection image on the target background image according to the lambert law, carrying out attenuation processing on the superposed image so as to output a final image obtained by the attenuation processing as a data set picture, wherein the target projection image is at least one of N intermediate projection images.
In a third aspect, there is provided an image processing apparatus comprising a memory, a processor, and an image processing program stored in the memory and running on the processor, the image processing program implementing the steps of the image processing method as in the first aspect.
In a fourth aspect, there is provided a computer storage medium which, when executed by a processor, carries out the steps of the image processing method as in the first aspect.
In a fifth aspect, a computer program product is provided, the computer program product comprising a computer program which, when executed by a processor, implements the steps of the image processing method as in the first aspect.
Compared with the prior art, the image processing method, the device, the equipment, the computer storage medium and the computer program product provided by the embodiment of the application have the advantages that the file obtained by modeling the target object is read, the modeling file is processed to obtain the simulation body data of the target object, the ray projection simulation is further carried out on the simulation body data, the influence of the energy spectrum curve of the ray source and the actual scattering and noise is comprehensively considered before and after the ray projection simulation, the data set picture finally obtained based on the lambert law is compared with the experimental image generated by the imaging system where the ray source is located, the physical design parameters and the principles are consistent, the image quality is close, and the image data set can be formed for wide use. On the other hand, by using the method, a large number of data set pictures can be obtained only by obtaining background images and modeling files, and the data set pictures are not required to be obtained through a large number of experiments, so that the time and the economic cost are effectively saved, and the technical problem that the data set pictures are manufactured with a large amount of time and economic cost is solved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort to a person of ordinary skill in the art.
Fig. 1 is an image processing method according to an embodiment of the present invention.
Fig. 2 is a detailed schematic flowchart of S120 of the image processing method of another embodiment.
Fig. 3 is an exemplary diagram of an initial projection image in an image processing method according to an embodiment of the present invention.
Fig. 4 is an exemplary diagram of a background image in an image processing method according to an embodiment of the present invention.
Fig. 5 is an exemplary diagram of a dataset picture in an image processing method according to an embodiment of the present invention.
Fig. 6 is a schematic block diagram of an image processing apparatus of an embodiment of the present invention.
Fig. 7 is a schematic block diagram of an image processing apparatus of an embodiment of the present invention.
Detailed Description
Features and exemplary embodiments of various aspects of the invention are described in detail below. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the invention by showing examples of the invention.
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The embodiments will be described in detail below with reference to the accompanying drawings.
As described in the background art, in the field of radiation image detection, in order to improve the speed of examination and the accuracy of target object recognition, deep learning techniques have been widely used in detection and recognition of radiation images. However, in practical application, the quality of the radiation image containing the target object is closely related to the design parameters of the imaging system applied by the radiation source, and the quality of images produced by different imaging systems often varies greatly, so that when the deep learning technology is applied to radiation image detection, the production of an experimental data set is subject to great objective constraint, the production is time-consuming and labor-consuming, and a large number of ideal radiation images containing the target object are difficult to obtain.
In addition, the number and quality of the data sets are critical to the algorithm based on machine learning, so in order to solve the technical problems, and ensure the good application of the machine learning algorithm to the radiation image detection, it is necessary to provide an effective image processing method, so that a large number of data set pictures meeting the actual conditions and requirements can be efficiently obtained, and the time and the economic cost are saved.
The target object may be prohibited articles by national laws and regulations, other articles, human beings, animals and plants, and the like. The imaging system of the radiation source application can be a security inspection system, which can be a security inspection system used in public transportation security operation, a large vehicle security inspection system and the like. The radiation source may be X-rays, or others.
The image processing method of the present application will be first described below.
Referring to fig. 1, in an embodiment of the image processing method of the present application, the method includes:
S110, reading a modeling file to obtain point-plane data of the target object, wherein the modeling file is obtained by modeling the target object.
S120, constructing simulation body data of the target object according to the material and the point-plane data of the target object and the energy spectrum curve of the ray source radiating the target object.
S130, performing ray projection simulation on the simulation body data of the target object to obtain N initial projection images, wherein N is a positive integer, and N Zhang Chushi projection images correspond to different target object arrangement positions.
S140, simulating scattering and noise in the actual imaging process, and performing blurring processing on the N Zhang Chushi projection images to obtain N intermediate projection images.
S150, superposing the target projection image on the target background image according to the lambert law, and carrying out attenuation processing on the superposed image so as to output a final image obtained by the attenuation processing as a data set picture, wherein the target projection image is at least one of N intermediate projection images.
According to the embodiment of the application, the file obtained by modeling the target object is read, the modeling file is processed to obtain the simulation body data of the target object, the ray projection simulation is further carried out on the simulation body data, and the energy spectrum curve of the ray source and the influence of actual scattering and noise are comprehensively considered before and after the ray projection simulation, so that the physical design parameters and principles are consistent, the image quality is close, and an image data set can be formed for wide use compared with an experimental image generated by an imaging system where the ray source is located in the final data set picture obtained based on the lambert law. On the other hand, by using the method, a large number of data set pictures can be obtained only by obtaining background images and modeling files, and the data set pictures are not required to be obtained through a large number of experiments, so that the time and the economic cost are effectively saved, and the technical problem that the data set pictures are manufactured with a large amount of time and economic cost is solved.
In an alternative embodiment, in S110, the modeling file may be a file obtained by three-dimensionally modeling the target object by using computer-aided design software, and the modeling file may be in STL (STereoLithography ) format. In modeling, it is necessary to ensure that the model passes the seal and validity test.
The modeling file is read by software to obtain the point-plane data of the target object, and the STL modeling file is still described by taking the modeling file in the STL format as an example, where the STL modeling file is a standard format file for storing triangular patches, and the point-plane data may include N tr triangular patches.
In an alternative example, in S120, the energy spectrum effect may be considered, and the point data is processed by using the energy spectrum curve of the radiation source radiating the target object and the materials at different positions of the target object, so as to construct the simulated volume data of the target object.
Referring to fig. 2, in an alternative example, the step S120 may include the steps of:
S210, performing space sampling and connected domain filling processing on the point-plane data of the target object to obtain binary data of the target object.
S220, performing material assignment on the binary volume data through the energy spectrum curve of the ray source radiating the target object and the material of the target object to obtain the simulation volume data of the target object.
In the example, an optional implementation process for constructing simulation body data based on an energy spectrum curve is provided, the basic idea is that binary body data of a target object is constructed, material assignment is carried out on the binary body data of corresponding areas of different materials in the target object according to attenuation degrees of the radiation source on the different materials according to energy spectrum effect, and when the material assignment is completed on the binary body data of all areas of the target object, the obtained simulation body data of the target object is obtained.
Optionally, the whole process of constructing binary data of the target object may include: performing space discrete sampling on the point-plane data of the target object to obtain a plurality of sampling points; traversing a plurality of sampling points, and assigning the volume data of the sampling points on the triangular patch as a first numerical value; and carrying out three-dimensional connected domain judgment on the sampling points which are not positioned on the triangular patch so as to assign the volume data of the sampling points which accord with the three-dimensional connected domain as a first value, and assign the volume data of the sampling points which do not accord with the three-dimensional connected domain as a second value, wherein the first value is not equal to the second value.
It should be noted that, when the point-plane data of the target object is spatially and discretely sampled to obtain a plurality of sampling points, the sampling points may be used to construct boundary data of the target object in the model. The number of the sampling points can be specifically set according to the requirement of actual simulation precision, and the higher the number of the sampling points is, the higher the fineness of the simulation model and the simulation image related to the final target object is.
The boundary data represents the boundary contour of the volume data of the target object, and the sampling points on the boundary data are pixel points on the boundary contour. After the boundary data of the target object is determined, the filling of the volume data of the target object can be continued to obtain three-dimensional region space data which accords with the model entity, namely binary volume data.
Still taking the modeling file in STL format as an example, the set of N tr triangular patches may be represented by the following formula (1).
Wherein Ω in the set represents one triangular patch in the point-plane data of the target object, and Ω outside the set represents the set of triangular patches in the point-plane data of the target object.
When constructing boundary data, the boundary data is set to any sampling pointWhen it meets/>I.e. any sampling point is on a triangular patch, the boundary data where the sampling point is located is assigned a first value, e.g. 1, i.e./>If the sampling point does not belong to all triangular patches, the boundary data where the sampling point is located may be temporarily assigned to a second value, where the first value and the second value are inconsistent, for example, the second value is 0. After traversing all the sampling points, boundary data of the target object can be obtained, and the boundary data can describe the boundary outline of the volume data of the target object.
After the boundary data of the target object is obtained, three-dimensional connected domain judgment can be performed on the boundary data, and if any sampling point which is not positioned on the boundary contour meets the definition of the three-dimensional connected domain, the three-dimensional connected domain can be marked asOtherwise, it is marked as/>In other words, finally, the volume data of the target object may be assigned and filled by the following formula (2), to obtain binary volume data VB of the target object, where the binary volume data VB characterizes the three-dimensional region space including the target object entity.
In an alternative example, the process of performing material assignment on binary data according to the energy spectrum effect may include: according to the energy spectrum curve of the ray source, determining the line attenuation coefficients of different materials in the target object when the ray source irradiates the target object; and carrying out material assignment on binary volume data of the target object area corresponding to different materials according to the line attenuation coefficients of the different materials to obtain simulation volume data of the target object.
In the above example, the line attenuation coefficient is assigned to the binary volume data in the divided regions to obtain the simulation volume data. Wherein the line attenuation coefficient can be determined according to the energy spectrum curve of the ray source and the materials of different areas of the target object.
The line attenuation coefficient may be an average attenuation coefficient calculated under the condition that the energy spectrum curve of the selected radiation source and the material of the target object are fixed. It can be understood that, for a single target object, the binary data can be divided into N spatial regions according to different materials, namely V 1、V2、V3、...、VN, each region corresponds to a different material M 1、M2、M3、...、MN, and the relationship between the linear attenuation coefficient and the energy corresponding to each material is thatThe energy spectrum curve and the average attenuation coefficient under the condition of fixed materials can be calculated according to the following formula (3).
Wherein,For the average attenuation coefficient, K is the energy segment number of the energy spectrum curve, f (E j) is the energy spectrum of the radiation source of the j-th segment, d is the physical size of the volume data pixel,/>Where S is the actual physical size of the target object in each direction, and N is the number of sampling points of the target object in each direction.
The following is an example of a target object as a contraband 0.380AP pistol. The contraband can be regarded as a single aluminum alloy material, and when the contraband is irradiated, the selected ray source is an isotope Co 60 source, and the energy spectrum curve of the isotope Co 60 source refers to the following formula (4):
when the line attenuation coefficient is calculated, the attenuation coefficient of the aluminum alloy under two energy sections can be taken, and the average attenuation coefficient can be calculated by referring to the following formula (5)
The physical dimension d of the volume data pixel in the above formula (5) takes a value of 1 mm.
After the line attenuation coefficient is obtained, material assignment can be carried out on the corresponding area of the binary volume data in a zoned mode, namely, the material of the binary volume data with the first value is assigned to the line attenuation coefficient calculated according to the material of the area and the spectral curve, and after the material assignment of all the areas is completed, the simulation volume data of the target object is obtained.
In some examples, in S130, a reasonable location space for placing the target object may be found according to the geometric layout of the imaging system, so as to select a placement mode of NL target objects, where these different placement modes may be adjusted and differentiated based on the central position vector diameter and the rotation angle of the simulated volume data.
The simulation body data can be subjected to ray projection simulation according to N different placement modes to obtain N initial projection images, and the final initial projection images can correspond to different target object placement positions as the ray projection simulation is based on different placement modes.
Note that, for NL different target object placement positions, the calculation formula of the obtained initial projection image may refer to the following formula (6).
imgPi＝∫lVS(Pi)·dl (6)
Wherein l is the connection line between the detector and the radiation source during the projection simulation of the radiation, and VS (P i) is to place the center of the volume data VS at the position r i and rotate by an angle θ i.
Still taking the target object as the prohibited article 0.380AP pistol for illustration, when the simulated body data of the pistol is subjected to ray projection simulation, various initial projection images can be produced through rotation and translation, please refer to fig. 3, which is a part of examples.
In some examples, in S140, when blurring the initial projection image while simulating scattering and noise in the actual imaging process, the following steps may be performed:
Respectively carrying out convolution blurring treatment on the N Zhang Chushi projection images; and adding noise to each initial projection image after convolution blurring processing to obtain N intermediate projection images.
For any of the initial projection images imgP i, the blurring process may be performed according to the following formula (7) to simulate scattering and noise in the actual image, so as to obtain a final simulation image imgPS i of the target object, i.e., an intermediate projection image.
imgPSi＝imgPi*kernel+noise (7)
Wherein, kernel is convolution kernel, which can take experience value according to the detection effect of the imaging system where the actual radiation source is located, noise is noise, and the noise can be Gaussian noise or Poisson noise.
In this embodiment, the fuzzy processing of the simulation image is realized by means of the convolution kernel and noise, so that the simulation image is more similar to the actual radiation imaging effect.
In some embodiments, before S150, M typical background images may be acquired, then photon count images, idle images, and zero point images corresponding to the M typical background images are experimentally measured, and then zero point idle correction is performed according to formula (8) and the experimentally measured images to obtain M background images.
Wherein imgBG i is the ith background image, imgBGC i is the photon count image of the ith typical background image, imgZero i is the zero point image of the ith typical background image, imgFull i is the no-load image of the ith typical background image.
Referring to fig. 4, when the system to which the radiation source is applied is a large vehicle security inspection system, an example of a background image obtained by no-load zero point correction can be referred to.
In some examples, in S150, after obtaining M background images and N intermediate projection images, when making a dataset picture at a time, one background image may be randomly selected as the target background image, and at least one intermediate projection image may be randomly selected as the target projection image, and then the at least one intermediate projection image (i.e., the target projection image) may be superimposed on the target background image according to lambert' S law. In addition, when the image is superimposed, the pixels are superimposed according to the corresponding positions. And then, carrying out exponential decay transformation to obtain a data set picture containing the target object, and completing data set construction when the data set picture obtained by accumulation reaches the set quantity, wherein radiation image identification based on a deep learning technology can be carried out based on the data set.
Refer to fig. 5, which is a portion of a dataset picture taken with a contraband 0.380AP pistol as the target object based on fig. 4.
The image processing method according to the embodiment of the present invention is described in detail above with reference to fig. 1 to 5, and the image processing apparatus according to the embodiment of the present invention will be described in detail below with reference to fig. 6.
Referring to fig. 6, in an embodiment, the image processing apparatus includes:
the reading module 610 may be configured to read a modeling file to obtain point-surface data of the target object, where the modeling file is a file obtained by modeling the target object;
A construction module 620, configured to construct simulated volume data of the target object according to the material and the point-plane data of the target object and an energy spectrum curve of a radiation source radiating the target object;
the simulation module 630 may be configured to perform ray projection simulation on the simulated volume data of the target object to obtain N initial projection images, where N is a positive integer, and N Zhang Chushi projection images correspond to different target object arrangement positions;
The blurring processing module 640 can be used for simulating scattering and noise in the actual imaging process, and blurring processing is performed on the N Zhang Chushi projection images to obtain N intermediate projection images;
The superposition attenuation module 650 may be configured to superimpose the target projection image on the target background image according to lambert law, and perform attenuation processing on the superimposed image, so as to output a final image obtained by the attenuation processing as a dataset picture, where the target projection image is at least one of the N intermediate projection images.
Alternatively, the building module may include:
The filling unit can be used for performing space sampling and connected domain filling processing on the point-plane data of the target object to obtain binary data of the target object;
And the material assignment unit can be used for carrying out material assignment on the binary volume data through the energy spectrum curve of the ray source radiating the target object and the material of the target object to obtain the simulation volume data of the target object.
Optionally, the point-plane data includes M triangular patches, M being a positive integer; the filling unit may include:
The sampling subunit can be used for carrying out space discrete sampling on the point-plane data of the target object to obtain a plurality of sampling points;
The traversing subunit can be used for traversing a plurality of sampling points and assigning the volume data of the sampling points on the triangular patch as a first numerical value;
The assignment subunit can be used for judging the three-dimensional connected domain of the sampling points which are not positioned on the triangular patch, so as to assign the volume data of the sampling points which accord with the three-dimensional connected domain to a first value, and assign the volume data of the sampling points which do not accord with the three-dimensional connected domain to a second value, wherein the first value is not equal to the second value.
Optionally, the material assigning unit may include:
the determining subunit can be used for determining the line attenuation coefficients of different materials in the target object when the radiation source irradiates the target object according to the energy spectrum curve of the radiation source;
And the material assignment subunit can be used for carrying out material assignment on binary volume data of the target object area corresponding to different materials according to the line attenuation coefficients of the different materials to obtain simulation volume data of the target object.
Optionally, the blurring processing module may include:
the convolution fuzzy processing unit can be used for respectively carrying out convolution fuzzy processing on the N Zhang Chushi projection images;
the noise adding unit can be used for adding noise to each initial projection image after convolution blurring processing to obtain N intermediate projection images.
Optionally, the image processing apparatus may further include:
the acquisition module can be used for acquiring M typical background images;
the correction module can be used for carrying out no-load zero point correction on M typical background images to obtain M background images; wherein the M background images include target background images.
Fig. 7 shows a schematic hardware configuration of an image processing apparatus according to an embodiment of the present application. The image processing device may include, among other things, a processor 701 and a memory 702 storing computer program instructions.
In particular, the processor 701 may comprise a Central Processing Unit (CPU), or an Application SPECIFIC INTEGRATED Circuit (ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present application.
Memory 702 may include mass storage for data or instructions. By way of example, and not limitation, memory 702 may include a hard disk drive (HARD DISK DRIVE, HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or universal serial bus (Universal Serial Bus, USB) drive, or a combination of two or more of the foregoing. The memory 702 may include removable or non-removable (or fixed) media, where appropriate. Memory 702 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 702 is a non-volatile solid state memory.
Memory 702 may include read-only memory (ROM), flash memory devices, random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, the memory 702 includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors) it is operable to perform the operations described with reference to the methods according to the above aspects of the disclosure.
The processor 701 implements any one of the image processing methods of the above embodiments by reading and executing computer program instructions stored in the memory 702.
In one example, the image processing device may also include a communication interface 703 and a bus 710. As shown in fig. 7, the processor 701, the memory 702, and the communication interface 703 are connected by a bus 710 and perform communication with each other.
The communication interface 703 is mainly used for implementing communication between each module, device, unit and/or apparatus in the embodiment of the present application.
Bus 710 includes hardware, software, or both that couple the components of the image processing device to each other. By way of example, and not limitation, the buses may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a micro channel architecture (MCa) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus, or a combination of two or more of the above. Bus 710 may include one or more buses, where appropriate. Although embodiments of the application have been described and illustrated with respect to a particular bus, the application contemplates any suitable bus or interconnect.
The image processing apparatus may be based on an image processing method, thereby implementing the image processing method and device described in connection with fig. 1 to 6.
In combination with the image processing method in the above embodiment, the embodiment of the present application may be implemented by providing a computer storage medium. The computer storage medium has stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement any of the image processing methods of the above embodiments.
In addition, in connection with the image processing method in the above embodiment, the embodiment of the present application may be implemented by providing a computer program product. The computer program product has computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement any of the image processing methods of the above embodiments.
In addition, the term "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.
It should be understood that in embodiments of the present invention, "B corresponding to a" means that B is associated with a, from which B may be determined. It should also be understood that determining B from a does not mean determining B from a alone, but may also determine B from a and/or other information.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (9)
1. An image processing method, comprising:
reading a modeling file to obtain point-plane data of a target object, wherein the modeling file is obtained by modeling the target object;
constructing simulation body data of the target object according to the material of the target object, the point-surface data and an energy spectrum curve of a ray source radiating the target object;
Performing ray projection simulation on the simulation body data of the target object to obtain N initial projection images, wherein N is a positive integer, and the N initial projection images correspond to different target object arrangement positions;
simulating scattering and noise in the actual imaging process, and carrying out fuzzy processing on N initial projection images to obtain N intermediate projection images;
Superposing a target projection image on a target background image according to the lambert law, and carrying out attenuation treatment on the superposed image so as to output a final image obtained by the attenuation treatment as a data set picture, wherein the target projection image is at least one of N intermediate projection images;
The constructing simulation body data of the target object according to the material of the target object, the point-surface data and the energy spectrum curve of a ray source radiating the target object comprises the following steps:
Performing space sampling and connected domain filling processing on the point-plane data of the target object to obtain binary data of the target object;
And carrying out material assignment on the binary volume data through the energy spectrum curve of the radiation source radiating the target object and the material of the target object to obtain the simulation volume data of the target object.
2. The method of claim 1, wherein the point-to-surface data comprises M triangular patches, M being a positive integer; the step of performing spatial sampling and connected domain filling processing on the point-plane data of the target object to obtain binary data of the target object comprises the following steps:
Performing space discrete sampling on the point-plane data of the target object to obtain a plurality of sampling points;
traversing a plurality of sampling points, and assigning the volume data of the sampling points on the triangular patch as a first numerical value;
And judging the three-dimensional connected domain of the sampling points which are not positioned on the triangular surface patch, so as to assign the volume data of the sampling points which accord with the three-dimensional connected domain to the first value, and assign the volume data of the sampling points which do not accord with the three-dimensional connected domain to the second value, wherein the first value is not equal to the second value.
3. The method according to claim 1, wherein the performing material assignment on the binary volume data by radiating the energy spectrum curve of the radiation source of the target object and the material of the target object to obtain the simulated volume data of the target object includes:
According to the energy spectrum curve of the ray source, determining the linear attenuation coefficients of different materials in the target object when the ray source irradiates the target object;
and carrying out material assignment on binary volume data of target object areas corresponding to different materials according to the line attenuation coefficients of the different materials to obtain simulation volume data of the target object.
4. The method according to claim 1, wherein said simulating the scattering and noise in the actual imaging process performs blurring processing on the N initial projection images to obtain N intermediate projection images, and includes:
respectively carrying out convolution blurring processing on N initial projection images;
And adding noise to each initial projection image after convolution blurring processing to obtain N intermediate projection images.
5. The method of claim 1, wherein the target projection image is superimposed on the background image prior to the superimposing, the method further comprising:
Obtaining M typical background images;
Carrying out zero-load correction on M typical background images to obtain M background images; wherein the M background images include the target background image.
6. An image processing apparatus, characterized in that the apparatus comprises:
the reading module is used for reading a modeling file to obtain point-plane data of a target object, wherein the modeling file is obtained by modeling the target object;
The construction module is used for constructing simulation body data of the target object according to the material of the target object, the point-surface data and the energy spectrum curve of a ray source radiating the target object;
The simulation module is used for carrying out ray projection simulation on the simulation body data of the target object to obtain N initial projection images, wherein N is a positive integer, and the N initial projection images correspond to different target object arrangement positions;
the fuzzy processing module is used for simulating scattering and noise in the actual imaging process, and performing fuzzy processing on the N initial projection images to obtain N intermediate projection images;
The superposition attenuation module is used for superposing a target projection image on a target background image according to the lambert law, carrying out attenuation treatment on the superposed image, and outputting a final image obtained by the attenuation treatment as a data set picture, wherein the target projection image is at least one of N intermediate projection images;
The constructing simulation body data of the target object according to the material of the target object, the point-surface data and the energy spectrum curve of a ray source radiating the target object comprises the following steps:
Performing space sampling and connected domain filling processing on the point-plane data of the target object to obtain binary data of the target object;
And carrying out material assignment on the binary volume data through the energy spectrum curve of the radiation source radiating the target object and the material of the target object to obtain the simulation volume data of the target object.
7. An image processing apparatus comprising a memory, a processor, and an image processing program stored in the memory and running on the processor, the image processing program performing the steps of the image processing method according to any one of claims 1 to 5.
8. A computer storage medium, characterized in that the computer storage medium, when executed by a processor, implements the steps of the image processing method according to any one of claims 1 to 5.
9. A computer program product, characterized in that the computer program product comprises a computer program which, when being executed by a processor, implements the steps of the image processing method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210322761.3A CN114758066B (en) | 2022-03-30 | 2022-03-30 | Image processing method, apparatus, device, storage medium, and computer program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210322761.3A CN114758066B (en) | 2022-03-30 | 2022-03-30 | Image processing method, apparatus, device, storage medium, and computer program product |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114758066A CN114758066A (en) | 2022-07-15 |
| CN114758066B true CN114758066B (en) | 2024-06-21 |
Family
ID=82327860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210322761.3A Active CN114758066B (en) | 2022-03-30 | 2022-03-30 | Image processing method, apparatus, device, storage medium, and computer program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114758066B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336272A (en) * | 2013-03-26 | 2013-10-02 | 中国科学院电子学研究所 | Geometric structure based complex target SAR image simulation method |
| CN106204673A (en) * | 2016-06-29 | 2016-12-07 | 中国人民解放军信息工程大学 | Based on energy spectral filter and the CT image metal artifact bearing calibration of Image Residual re-projection |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102015207107A1 (en) * | 2015-04-20 | 2016-10-20 | Siemens Healthcare Gmbh | Method for generating a virtual X-ray projection on the basis of an image data set obtained by means of an X-ray image recording device, computer program, data carrier and X-ray image recording device |
| CN108230286B (en) * | 2016-12-16 | 2022-01-11 | 北京机电工程研究所 | Method for rapidly analyzing influence of environmental and time-period factors on infrared detectability of target |
| CN112348915A (en) * | 2020-11-16 | 2021-02-09 | 上海联影医疗科技股份有限公司 | X-ray system simulation method, system, readable storage medium and device |
-
2022
- 2022-03-30 CN CN202210322761.3A patent/CN114758066B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336272A (en) * | 2013-03-26 | 2013-10-02 | 中国科学院电子学研究所 | Geometric structure based complex target SAR image simulation method |
| CN106204673A (en) * | 2016-06-29 | 2016-12-07 | 中国人民解放军信息工程大学 | Based on energy spectral filter and the CT image metal artifact bearing calibration of Image Residual re-projection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114758066A (en) | 2022-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220035961A1 (en) | System and method for artifact reduction of computed tomography reconstruction leveraging artificial intelligence and a priori known model for the object of interest | |
| CN104024792B (en) | Tire shape inspection method and tire shape inspection device | |
| US8345949B2 (en) | Sequential approach for automatic defect recognition | |
| US8774482B2 (en) | Generating pseudo-CT image volumes from ultra-short echo time MR | |
| CN107609483B (en) | Dangerous target detection method and device for driving assistance system | |
| Tyystjärvi et al. | Automated defect detection in digital radiography of aerospace welds using deep learning | |
| US10360670B2 (en) | Method for the non-destructive testing of the volume of a test object and testing device configured for carrying out such a method | |
| EP4006834A1 (en) | Damage detection using machine learning | |
| CN111553858A (en) | Image restoration method and system based on generation countermeasure network and application thereof | |
| JP2021530050A (en) | Article inspection by radiation imaging using repeated projection matching method | |
| CN113592886B (en) | Drawing examining method and device for building drawing, electronic equipment and medium | |
| CN114693660B (en) | ICT-based solid rocket engine charge calculation grid generation method | |
| CN107993201B (en) | Depth image enhancement method with retained boundary characteristics | |
| CN113971669A (en) | Three-dimensional detection system applied to pipeline damage identification | |
| CN115249321A (en) | Method for training neural network, system for training neural network and neural network | |
| CN114758066B (en) | Image processing method, apparatus, device, storage medium, and computer program product | |
| CN114581605A (en) | Method, device and equipment for generating scanning image of workpiece and computer storage medium | |
| Tsukahara et al. | Coupled tomography and distinct-element-method approach to exploring the granular media microstructure in a jamming hourglass | |
| Mery et al. | Simulation of defects in aluminium castings using CAD models of flaws and real X-ray images | |
| CN116051541B (en) | Bearing end face gentle abrasion detection method and device based on stroboscopic light source | |
| CN114581433B (en) | Method and system for acquiring surface morphology detection image in metal ball cavity | |
| Ziabari et al. | Simurgh: A Framework for CAD-Driven Deep Learning Based X-Ray CT Reconstruction | |
| JP6956195B2 (en) | How to detect multiple edges and devices | |
| US9064307B2 (en) | Methods and apparatus to generate three-dimensional spinal renderings | |
| Huang et al. | A template model for defect simulation for evaluating nondestructive testing in X-radiography |
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 | ||
| GR01 | Patent grant |