Method for 3D image reconstruction
11160521 · 2021-11-02
Assignee
Inventors
Cpc classification
A61B6/027
HUMAN NECESSITIES
A61B6/5229
HUMAN NECESSITIES
A61B6/5223
HUMAN NECESSITIES
A61B6/5205
HUMAN NECESSITIES
International classification
Abstract
A method for 3D image reconstruction includes the steps of radiating an X-ray source at a predetermined angle to a photographing subject and acquiring a plurality of two-dimensional image images received through a detector, generating an image population by scaling the obtained plurality of two-dimensional image images, generating a virtual image using a pixel average between two-dimensional images from the plurality of scale-corrected two-dimensional images, and adding the virtual image to the image population, repeatedly generating a virtual image using a pixel average between two-dimensional images included in the image population until a two-dimensional image larger than a predetermined number P is generated in the image population, and generating a 3D video image using the generated 2D video image.
Claims
1. A method for three-dimensional (3D) image reconstruction, the method comprising: radiating an X-ray source at a predetermined angle to a photographic subject and obtaining a plurality of two-dimensional images received through a detector; generating an image population by scale correcting the obtained plurality of two-dimensional images; generating a virtual image from the plurality of scale-corrected two-dimensional images using a pixel average between the two-dimensional images and adding the virtual image to the image population; repeating the generating of the virtual image using a pixel average between two-dimensional images included in the image population until a number of two-dimensional images are generated in the image population more than a predetermined number P; and generating a 3D video image using the generated two-dimensional images.
2. A method of claim 1, wherein the obtaining of the plurality of two-dimensional images comprises obtaining a vertical image produced by the x-ray source radiated vertically towards the detector, and a first side image and a second side image respectively generated by an X-ray source radiated in a predetermined angularly rotated direction in a first direction and a second direction based on the vertical image.
3. A method of claim 2, wherein the generating of the image population by the scale correction is characterized in that the first side image and the second side image are scale corrected to the size of the vertical image.
4. A method of claim 2, wherein the repetitively performing is characterized in that the virtual image is generated over a plurality of times, wherein 2n virtual images are generated at the time of nth generation, and the virtual image is repeatedly generated as many times as n corresponding to n calculated by Equation 1:
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
DETAILED DESCRIPTION
(9) Hereinafter, with reference to the accompanying drawings will be described in more detail a method for 3D image reconstruction according to an embodiment of the present invention. However, the present invention is not limited by the embodiments. Same reference numerals in the drawings denote same elements.
(10) In an embodiment of the present invention, there is provided a method that in the process of inspecting the inside of a subject through an X-ray inspection apparatus, a two-dimensional image of various angles may be reconstructed into a three-dimensional image, and a three-dimensional image may be configured using a minimum number of two-dimensional images.
(11) Hereinafter, the 3D image reconstruction method will be described in more detail.
(12)
(13) First, referring to
(14) As shown in
(15) Meanwhile, referring to
(16) In this case, in the case of the first side image 242 and the second side image 243, when the incident angle of the X-ray source emitted from the X-ray radiator 230 and the receiving angle of the detector are perpendicular (231), the distortion of the image will not occur. However, if the incident angle and the receiving angle are not perpendicular (232), image distortion occurs.
(17) Therefore, in the case of the embodiment of the present invention, which emits the X-ray source toward the planar detector at various angles, because the incident angle and the receiving angle are not perpendicular to each other, the X-ray source radiating from the side may cause distortion of the image.
(18) In the step 120, an image population may be generated by scaling the obtained plurality of 2D image images 241, 242, and 243.
(19) In this case, in step 120, the first side image 242 and the second side image 243 may be scale-corrected to the size of the vertical image 241. Since the vertical image 241 is an area where the X-ray source is vertically incident, there is no distortion of the image, it is preferable to perform scaling correction based on this.
(20) Referring to
(21) In the step 130, a virtual image using a pixel average between two-dimensional images may be generated from the plurality of scale-corrected two-dimensional images and added to the image population.
(22) More specifically, as shown in
(23) In a similar way, the second virtual image 260 is generated between the vertical image 241 and the second corrected image 2431 using the pixel average between the vertical image 241 and the second corrected image 2431, and the generated second virtual image 260 can also be added to the image population.
(24) So, in this case, two virtual images 250 and 260 generated using pixel averages are added to three two-dimensional images 241, 242 and 243 generated through X-ray imaging, and thus five images 241, 242, 243, 250, and 260 may be stored in the image population.
(25) Thereafter, in step 140, the virtual image generation is repeatedly performed by using a pixel average between two-dimensional images included in the image population until a number of two-dimensional images larger than a predetermined number P are generated in the image population.
(26) That is, in the above example, by generating the first virtual image using the pixel average, the first virtual image and the second virtual image are generated, and five image images 241, 242, 243, 250, and 260 are sequentially stored in the image population. In addition, when the second virtual image generation operation is performed, virtual images may be generated between five image images, and thus four virtual images may be additionally generated.
(27) In more detail, when the number of photographed 2D video images is M (ex. 3), the number of generated virtual images is (M−1)×2.sup.n-1 each time the virtual image generating operation is performed n times, and thus by repeatedly performing the virtual image generation until the cumulative sum of the generated virtual images is larger than a predetermined number, it is possible to generate as many 2D images as necessary to reconstruct the 3D image.
(28) In this way, since the total number of virtual images that can be generated by repeating n virtual image creation operations is
(29)
the generating of the virtual image may be performed until the total sum of the virtual images is equal to or greater than a predetermined number P. This may be expressed as follows by Equation 1 below.
(30)
(31) where P is a predetermined number
(32) Thereafter, in step 150, a 3D video image is generated using the generated 2D video image.
(33) Therefore, by comparing the generated three-dimensional image with the good image it can be more accurately determine whether the product is defective.
(34) Meanwhile, in one embodiment of the present invention, the X-ray inspection apparatus may be used in various forms.
(35) Referring to
(36) Accordingly, there are four two-dimensional images generated from four X-ray sources, and a virtual image using pixel averages may be generated from the four images. That is, three images are generated through the first virtual image generation, the virtual images generated through the second virtual image generation are 3×2=6, and the virtual image generated through the nth virtual image generation may be
(37)
as shown in Equation 1 above.
(38) Therefore, even in this case, the virtual image generating operation may be repeated until the total sum of the generated virtual images is equal to or greater than the predetermined number P, thereby sufficiently generating the two-dimensional image for the three-dimensional image reconstruction.
(39) As described above, according to an embodiment of the present invention, after acquiring a small number of 2D images, by generating a virtual image using the acquired image and using it to generate a 3D image, a 3D image reconstruction method capable of minimizing the exposure of radiation through at least 2D imaging may be provided.
(40) In addition, according to an embodiment of the present invention, a three-dimensional image reconstruction method capable of realizing a three-dimensional image by minimizing image distortion may be provided by correcting the image scale between two-dimensional images taken from multiple angles and using them in three-dimensional image reconstruction.
(41) In addition, the method for 3D image reconstruction according to an embodiment of the present invention may be recorded in a computer readable medium including program instructions for performing various computer-implemented operations. The computer readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions on the medium may be those specially designed and constructed for the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include hard disks, floppy disks, magnetic media such as magnetic tape, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by the compiler, but also high-level language code that can be executed by a computer using an interpreter.
(42) As described above, although one embodiment of the present invention has been described with reference to a limited embodiment and drawings, the present invention is not limited to the above-described embodiment. It is possible for those skilled in the art to various modifications and variations from this description. Accordingly, the present invention should be understood only by the claims set forth below, and all equivalent or equivalent modifications thereof will fall within the scope of the present invention.