Method, device, medium and equipment for determining the number of slice layers of a three-dimensional image

Abstract

The present invention discloses a method, device, medium, and equipment for determining the number of slice layers in a three-dimensional image, relating to the field of three-dimensional display imaging technology. The method includes: determining the number of left neighboring slice layers N‑n and the number of right neighboring slice layers N+n based on the initial number of slice layers N and the initial search range n; obtaining restored three-dimensional images corresponding to the number of layers N, N‑n, and N+n, capturing two-dimensional plane images of the restored three-dimensional image from different angles, and obtaining characteristic values ​​for characterizing the quality of the restored three-dimensional image based on the pixel distribution in the two-dimensional plane images at different angles; using the number of slice layers with the optimal characteristic value as the new initial number of slice layers, and repeating the operation until the optimal number of slice layers is obtained. By analyzing the pixel distribution in the two-dimensional plane images of the restored three-dimensional image captured at different angles, the quality of the restored three-dimensional image can be quantified more comprehensively and accurately, thereby obtaining the number of slices that maximizes the imaging mechanical efficiency and achieves the best imaging effect.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional display imaging technology, and in particular to a method, device, medium and equipment for determining the number of slice layers of a three-dimensional image. Background Art

[0002] Slice-based reconstruction techniques first scan the object to be displayed using computed tomography (CT) or other 3D scanning technologies to obtain 3D data. A 3D model is then constructed from this data. Software is then used to segment the 3D model into a series of parallel 2D slices, each representing a cross-section of the object at a specific depth. Electronic display technology, such as a projector, is then used to sequentially project the 2D slices onto a "screen" at different depths to reconstruct the 3D model. This method has widespread application in fields such as geological exploration and industrial design. The resolution of the reconstructed 3D model in the slice projection direction is determined by the number of slice layers. With a small number of slice layers, the reconstructed image is stable but has low resolution, failing to better capture the details and features of the 3D model. With a large number of slice layers, the image resolution increases, but the image becomes unstable due to the mechanical frequency limits of the scanning mirror. Therefore, determining the optimal number of slice layers is crucial for the resolution of the reconstructed 3D model.

[0003] Since the restored three-dimensional image is projected onto a "screen" at different depths, it is difficult to obtain the voxels (the smallest unit that makes up the three-dimensional image) in the restored three-dimensional image, and therefore it is difficult to quantify the quality of the restored three-dimensional image. As a result, the current determination of the number of slice layers is mainly based on human eye observation and analysis of the imaging quality of the restored three-dimensional image. The operator decides the number and position of each slice by directly observing the clarity of the restored three-dimensional image.

[0004] However, the subjectivity of human observation makes it difficult to accurately evaluate the restoration quality of three-dimensional images. Different operators may obtain inconsistent slicing results on the same data, making it difficult to maximize the efficiency of the imaging machine while ensuring the imaging effect. Summary of the Invention

[0005] Based on this, it is necessary to provide a method, device, medium and equipment for determining the number of slice layers of a three-dimensional image to address the above technical problems.

[0006] This manual adopts the following technical solutions:

[0007] This specification provides a method for determining the number of slice layers of a three-dimensional image, comprising:

[0008] Determine the initial slice layer number N and the initial search range n, where both N and n are positive integers. Based on the initial slice layer number N and the initial search range n, determine the left neighboring slice layer number Nn and the right neighboring slice layer number N+n.

[0009] Restored 3D images corresponding to the initial slice layer number N, the left neighboring slice layer number Nn, and the right neighboring slice layer number N+n are obtained respectively, and 2D plane images corresponding to the restored 3D images are collected from different angles. Based on the pixel connectivity or edge pixel density in the 2D plane images at different angles, characteristic values ​​used to characterize the quality of each restored 3D image are obtained;

[0010] The number of slice layers corresponding to the optimal eigenvalue is used as the new initial number of slice layers, and the initial search range is narrowed; the eigenvalue acquisition operation is repeated according to the new initial number of slice layers and the reduced initial search range until the fluctuation of the eigenvalue is less than the set threshold, and the optimal number of slice layers is obtained.

[0011] Furthermore, the characteristic value includes a connectivity characteristic value P1 for characterizing the connectivity of voxels in the three-dimensional image, specifically including:

[0012] Collect two-dimensional plane images of the restored three-dimensional image corresponding to the slice layer number i from different angles;

[0013] Use the connected domain analysis method to obtain the number of connected pixel sets in each two-dimensional plane image;

[0014] Take the weighted sum of the number of connected pixel sets to obtain the eigenvalue P1 corresponding to the slice layer number i:

[0015] P1=(n 1 +n 2 +…+n J ) / J

[0016] Among them, n 1 ~n J are the numbers of connected pixel sets in the J two-dimensional plane images obtained from J angles respectively. The above formula means that the quality of the restored three-dimensional image is characterized by the number of connected pixel sets at different angles: the larger the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the worse the restoration quality; the smaller the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the better the restoration quality.

[0017] Furthermore, the eigenvalues ​​include edge eigenvalues ​​P2 for characterizing the clarity of edges of the three-dimensional image, specifically including:

[0018] Collect two-dimensional plane images of the restored three-dimensional image corresponding to the slice layer number i from different angles;

[0019] Obtain edge pixel density in each two-dimensional plane image based on an edge detection algorithm;

[0020] The weighted sum of all edge pixel densities is used to obtain the eigenvalue P2 corresponding to the slice layer number i:

[0021] P2=(m 1 +m 2 +…+m J ) / J

[0022] Among them, m 1 ~m J are the edge pixel densities in the J two-dimensional plane images obtained from J angles respectively. The above formula means that the quality of the restored three-dimensional image is characterized by the edge pixel density at different angles: the greater the edge pixel density in each two-dimensional plane image, the clearer the edge in the restored three-dimensional image and the better the restoration quality; the smaller the edge pixel density in each two-dimensional plane image, the blurrier the edge in the restored three-dimensional image and the worse the restoration quality.

[0023] Furthermore, the restored three-dimensional image is obtained by scanning the shaped linear light source in the display material using a scanning galvanometer to form a continuous multi-layer screen, and projecting different sections of the three-dimensional model onto different screens in sequence through a digital micromirror display.

[0024] Furthermore, the wavelength of the linear light source is 1550 nm, and the curtain is a single-frequency excited curtain formed along the path when the linear light source passes through tellurite glass doped with trivalent erbium ions Er3+.

[0025] Furthermore, the digital micromirror display projects a pattern formed by irradiating a light source with a wavelength of 850 nm onto tellurite glass doped with trivalent erbium ions Er3+, thereby forming a dual-frequency excitation to generate a green pattern on the screen.

[0026] A device for determining the number of slice layers of a three-dimensional image, comprising:

[0027] A search initialization module is used to determine the initial number of slice layers N and the initial search range n, where both N and n are positive integers. Based on the initial number of slice layers N and the initial search range n, the number of left neighboring slice layers Nn and the number of right neighboring slice layers N+n are determined.

[0028] A quality evaluation module is used to obtain restored 3D images corresponding to the initial slice layer number N, the left neighboring slice layer number Nn, and the right neighboring slice layer number N+n, respectively, collect 2D plane images corresponding to the restored 3D images from different angles, and obtain characteristic values ​​used to characterize the quality of each restored 3D image based on the pixel connectivity or edge pixel density in the 2D plane images at different angles;

[0029] The module for determining the number of slice layers is used to use the number of slice layers corresponding to the optimal eigenvalue as the new initial number of slice layers and narrow the initial search range; the eigenvalue acquisition operation is repeated according to the new initial number of slice layers and the reduced initial search range until the fluctuation of the eigenvalue is less than the set threshold, thereby obtaining the optimal number of slice layers.

[0030] This specification provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method for determining the number of slice layers of a three-dimensional image is implemented.

[0031] This specification provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned method for determining the number of slice layers of a three-dimensional image is implemented.

[0032] At least one of the above technical solutions adopted in this specification can achieve the following beneficial effects:

[0033] The present invention can obtain a more comprehensive multi-angle view of the restored three-dimensional image by collecting two-dimensional plane images corresponding to the restored three-dimensional image from different angles. By analyzing the pixel distribution in the two-dimensional plane images at different angles, the restoration status of the three-dimensional image at different angles can be quantified, thereby more comprehensively and accurately quantitatively evaluating the quality of the restored three-dimensional image. Based on the evaluation results, the number of slice layers is adjusted to obtain the number of slices that maximizes the mechanical efficiency of the imaging and achieves the best imaging effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0035] Figure 1 A flow chart of a method for determining the number of slice layers of a three-dimensional image provided in this specification;

[0036] Figure 2 This is the principle diagram provided in this manual for restoring a three-dimensional model using two-dimensional slices of the three-dimensional model;

[0037] Figure 3 This is the principle diagram provided in this manual for restoring a cube with a side length of M from N layers of two-dimensional slices;

[0038] Figure 4 This is the principle diagram provided in this specification for restoring a cube with a side length of M from Nn layers of two-dimensional slices;

[0039] Figure 5This is the principle diagram provided in this manual for restoring a cube with a side length of M from N+n layers of two-dimensional slices;

[0040] Figure 6 The following are two-dimensional planes corresponding to different slices of a cube acquired from the same angle using volumetric three-dimensional imaging technology, as provided in this specification; (a), (b), (c), (d), and (e) correspond to the imaging results of the 15th, 20th, 25th, 30th, and 35th slices of the cube, respectively;

[0041] Figure 7 are the projected vertical stripes and horizontal stripes provided in this specification; (a) and (b) are vertical stripes and horizontal stripes with a resolution of 1280×800; (c) and (d) are vertical stripes and horizontal stripes with a resolution of 800×25;

[0042] Figure 8 This is a schematic diagram of imaging based on the dual-frequency up-conversion principle and using a scanning galvanometer and a digital micromirror device, as provided in this manual.

[0043] Description of reference numerals:

[0044] 1-digital micromirror device, 2-imaging lens group, 3-projection light, 4-trillion-doped tellurite glass, 5-linear light source, 6-scanning galvanometer. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of this specification more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments of this specification and the corresponding drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this invention.

[0046] Example 1

[0047] The method provided in this embodiment is based on the principle of restoring a 3D model using volumetric 3D display slices, optimizing slice parameters that affect image resolution and brightness. This method can be used in fields such as true 3D display imaging. True 3D display technology has a wide range of applications in military simulation, medical imaging, industrial design, flight simulators, advertising, film, and exhibitions. For many applications, factors that affect the restoration effect include image resolution and brightness. Due to current technological limitations, imaging quality has reached a bottleneck. Based on this, this embodiment optimizes the slice parameters of current volumetric 3D display imaging to improve imaging quality.

[0048] True 3D display technology, such as static volumetric 3D display based on dual-frequency upconversion, has the following imaging characteristics: First, it utilizes a dual-frequency infrared light source to generate visible green light, while invisible infrared light does not affect the display during the imaging process. Second, the display material must have excellent transparency and luminous efficiency. Third, the imaging method uses devices such as scanning galvanometers and digital micromirror display systems to restore the 3D model by taking 2D slices of the 3D model.

[0049] Imaging methods employ a dual galvanometer approach, focusing two light beams onto the same point. This produces high-brightness voxels, but due to mechanical frequency limitations, the resulting image resolution within the display material is low and the number of voxels is small, making it difficult to display complex images. Imaging using two digital micromirror displays distributes the energy of a single light spot across a two-dimensional plane. While this produces a large number of voxels, the brightness of the resulting three-dimensional image is very low due to the low luminous efficiency of the luminescent material. Combining the advantages of both approaches, a scanning galvanometer coupled with a digital micromirror display is employed. The scanning galvanometer rapidly scans a shaped linear light source across the display material, forming a continuous "curtain." The digital micromirror display then projects successive slices of the 3D model onto these different "curtains" to restore the 3D model. The resolution of one dimension of this method depends on the number of slices. Theoretically, for the same display volume, more slices provide better clarity and resolution. Current slicing methods include continuous parallel plane slicing and rotating plane slicing, but neither approach has explored the number of slices used.

[0050] Based on this, this embodiment provides a method for improving the resolution of static volumetric 3D display. When restoring a 3D model using a slicing method, the resolution of a certain dimension is determined by the number of slices. This embodiment, based on the imaging principle, maximizes the mechanical maximum frequency to find the maximum number of slice layers for stable 3D image display, thereby improving the volumetric 3D display resolution.

[0051] Figure 1 The flow chart of the method for determining the number of slice layers of a three-dimensional image according to this embodiment is shown. Figure 1 The method for determining the number of slice layers of a three-dimensional image is described in detail, and specifically comprises the following steps:

[0052] S101: Determine the initial number of slice layers N and the initial search range n, where both N and n are positive integers. Based on the initial number of slice layers N and the initial search range n, determine the number of left neighboring slice layers Nn and the number of right neighboring slice layers N+n.

[0053] The principle of restoring a 3D model by rapid sequential projection of 2D slices is as follows Figure 2As shown. The resolution in the XZ plane is determined by the digital microcrystal device, while the resolution in the YZ plane is determined by the number of slice layers. In theory, for the same display volume, the more layers there are, the higher the resolution. The principle of improving resolution by optimizing the number of slice layers is as follows Figure 3 、 4 , as shown in 5.

[0054] Figure 3 For example, a cube with a side length of M is restored by slicing N layers. When the initial number of slice layers is N, the spacing between each slice layer is L1 = M / N. The imaging results of different numbers of slice layers are performed on the left and right sides of layer N for comparison. It should be noted that since the target volume displayed by the imaging does not change, when the number of slice layers changes, the corresponding spacing between each layer also changes. For example Figure 4 As shown, when the number of layers is Nn, the corresponding interlayer spacing is L2 = M / Nn. Similarly, when the model is sliced ​​into N+n layers, the corresponding interlayer spacing is L3 = M / N+n, as shown in Figure 5 shown.

[0055] S102: Obtain restored 3D images corresponding to the initial slice layer number N, the left neighboring slice layer number Nn, and the right neighboring slice layer number N+n, respectively, collect 2D plane images of the restored 3D images from different angles, and obtain feature values ​​for characterizing the quality of the restored 3D images based on the pixel distribution in the 2D plane images at different angles.

[0056] The eigenvalue includes a connectivity eigenvalue P1 used to characterize the connectivity of voxels in the three-dimensional image. Specifically, the method includes: collecting two-dimensional plane images of the restored three-dimensional image corresponding to slice layer number i from different angles; using a connected domain analysis method to obtain the number of connected pixel sets in each two-dimensional plane image; and taking a weighted sum of the numbers of all connected pixel sets to obtain the eigenvalue P1 corresponding to slice layer number i:

[0057] P1=(n 1 +n 2 +…+n J ) / J

[0058] Among them, n 1 ~n J are the numbers of connected pixel sets in the J two-dimensional plane images obtained from J angles respectively. The above formula means that the quality of the restored three-dimensional image is characterized by the number of connected pixel sets at different angles: the larger the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the worse the restoration quality; the smaller the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the better the restoration quality. Figure 6 The 2D planes corresponding to cubes of different layers collected from the same angle are displayed using volumetric 3D imaging technology. Figure 6(a), (b), (c), (d), and (e) correspond to the imaging results of 15, 20, 25, 30, and 35 slices of the cube graphic, respectively. When the cube is divided into only 15 layers, the image is incomplete, resulting in a significant loss of detail. When the number of layers is increased to 20, the effect is slightly improved, but some graphic features are still missing. At 30 layers, the image begins to flicker, and when the number of layers reaches 35, the flickering phenomenon intensifies. This is due to the scanning frequency limitations of hardware such as the scanning galvanometer and projection system, which causes the displayed image to flicker and be unstable. Experimental data shows that 25-layer segmentation provides the most favorable balance, the pixels in the resulting image are the least dispersed, and the restoration quality is the best among them.

[0059] The above-mentioned eigenvalues ​​also include an edge eigenvalue P2 used to characterize the edge clarity of the 3D image. Specifically, the method includes: collecting 2D plane images of the restored 3D image corresponding to slice layer number i from different angles; obtaining the edge pixel density in each 2D plane image based on the edge detection algorithm; and performing a weighted summation of all edge pixel densities to obtain the eigenvalue P2 corresponding to slice layer number i:

[0060] P2=(m 1 +m 2 +…+m J ) / J

[0061] Among them, m 1 ~m J are the edge pixel densities in the J two-dimensional plane images obtained from J angles respectively. The meaning of the above formula is to characterize the quality of the restored three-dimensional image by the edge pixel density at different angles: the greater the edge pixel density in each two-dimensional plane image, the clearer the edge in the restored three-dimensional image and the better the restoration quality; the smaller the edge pixel density in each two-dimensional plane image, the blurrier the edge in the restored three-dimensional image and the worse the restoration quality. The resolution is estimated by projecting horizontal stripes and vertical stripes to determine the minimum spacing required to distinguish the two stripes. When a microelectronic device DMD with a resolution of 1280×800 is used for projection and the number of slice layers is 25, the voxel volume of the volumetric three-dimensional display system is 1280×800×25. Vertical and horizontal stripes are projected on a plane with a resolution of 1280×800, and photos are taken to record the photos that can just distinguish the two stripes. Figure 7 (a) and Figure 7 (b) in the figure shows vertical and horizontal stripes with a resolution of 1280×800, which determines that the resolution in the vertical and horizontal directions is about 300 microns. Then, the vertical and horizontal stripes are projected onto a plane with a resolution of 800×25. Figure 7 (c) and Figure 7(d) in the figure shows vertical and horizontal stripes with a resolution of 800×25, and it is determined that the resolution in the vertical and horizontal directions is approximately 400 μm and 300 μm, respectively.

[0062] S103: The number of slice layers with the best eigenvalue is used as the new initial number of slice layers, and the initial search range is reduced; the operation is repeated according to the new initial number of slice layers and the reduced initial search range until the fluctuation of the eigenvalue is less than the set threshold, and the optimal number of slice layers is obtained.

[0063] By observing the display effects under three different numbers of slice layers, if the display effect is most stable and clear relative to the Nn layer and the N+n layer when the number of layers is N, then continue to narrow the range on the left and right sides of the N layer to find the most stable and clear slice layer; if the Nn layer or the N+n layer is the most stable and clear, continue to search in the same way, and finally find the slice layer with the largest number of layers that can produce clear and stable imaging results to ensure higher resolution.

[0064] When the number of layers is less than the optimal number, the image effect is stable, but the resolution is low, and the details and features of the 3D model cannot be better displayed; when the number of layers is more than the optimal number, the image is unstable and unclear because the mechanical frequency of the scanning galvanometer has reached its limit and cannot produce higher frequency scanning.

[0065] In summary, this embodiment is based on a dual-optical path imaging system of a scanning galvanometer and a digital micromirror device, which optimizes the number of slice layers in the process of restoring a three-dimensional model from two-dimensional slices, so as to ensure the resolution of the three-dimensional image with as many slice layers as possible.

[0066] Furthermore, in one or more embodiments of the present specification, the three-dimensional image is restored by using a scanning galvanometer to scan a shaped linear light source within a display material to form a continuous multi-layer screen, and then projecting different sections of the three-dimensional model onto different screens in sequence through a digital micromirror display.

[0067] In the field of volumetric 3D display, static volumetric 3D displays based on dual-frequency upconversion offer advantages such as stable and clear imaging and a stable and flexible structure. Tellurite glass doped with trivalent erbium ions (Er3+) absorbs two infrared beams at wavelengths of 850nm and 1550nm, respectively, emitting green fluorescence at a wavelength of 546nm. Applying the principle of dual-frequency upconversion to volumetric 3D display, the two infrared beams intersect in the erbium-doped tellurite glass, generating visible green fluorescence at the intersection point, which serves as the voxel of the volumetric 3D display. This is then combined with a dual-optical imaging method using a scanning galvanometer and a digital micromirror device for imaging.

[0068] like Figure 8As shown, a 1550nm wavelength light source is shaped into a linear light source. This linear light source passes through tellurite glass, forming a single-frequency excitation "curtain" along its path. An 850nm wavelength light source is then applied to a digital microcrystal display device, projecting the pattern on the device into the tellurite glass. This dual-frequency excitation creates a visible green pattern on the "curtain." A scanning galvanometer rapidly scans the tellurite glass, creating a continuous multi-layer "curtain." Simultaneously, the digital micromirror device sequentially projects 2D slices of different layers of the 3D model onto the "curtain," reconstructing the 3D model.

[0069] Therefore, the number of voxels in the final 3D image is related to the number of slice layers and the parameters of the DMD. For example, if the number of slice layers is N and the DMD parameters are 1280×800, the number of voxels is N×1280×800. Therefore, the more slice layers there are, the more voxels there are, and the higher the 3D image resolution. This manual maximizes the resolution of volumetric 3D displays by optimizing the number of slice layers.

[0070] The above is a method for determining the number of 3D image slice layers provided in one or more embodiments of this specification. Based on the same idea, this specification also provides a corresponding device for determining the number of 3D image slice layers, including:

[0071] The search initialization module is used to determine the initial number of slice layers N and the initial search range n, where N and n are both positive integers. Based on the initial number of slice layers N and the initial search range n, the number of left neighboring slice layers Nn and the number of right neighboring slice layers N+n are determined.

[0072] The quality evaluation module is used to obtain the restored three-dimensional images corresponding to the initial slice layer number N, the left neighbor slice layer number Nn, and the right neighbor slice layer number N+n, respectively, collect two-dimensional plane images of the restored three-dimensional image from different angles, and obtain characteristic values ​​used to characterize the quality of the restored three-dimensional image based on the pixel distribution in the two-dimensional plane images at different angles.

[0073] The slice layer number determination module is used to use the slice layer number with the optimal eigenvalue as the new initial slice layer number and reduce the initial search range; repeat the operation according to the new initial slice layer number and the reduced initial search range until the fluctuation of the eigenvalue is less than the set threshold, thereby obtaining the optimal number of slice layers.

[0074] The specific limitations of the apparatus for determining the number of 3D image slice layers can be found in the limitations of the method for determining the number of 3D image slice layers described above and will not be further elaborated here. Each module in the apparatus for determining the number of 3D image slice layers described above may be implemented in whole or in part via software, hardware, or a combination thereof. Each of the modules may be embedded in or independent of a processor in a computer device in hardware form, or may be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0075] This specification also provides a computer-readable storage medium, which stores a computer program. The computer program can be used to execute the above-mentioned method for determining the number of three-dimensional image slice layers.

[0076] This specification also provides the structure of a computer device. At the hardware level, the computer device includes a processor, an internal bus, a network interface, memory, and non-volatile storage, and may also include other hardware required for operations. The processor reads the corresponding computer program from the non-volatile storage into the internal memory and then executes it to implement the method for determining the number of slice layers in a three-dimensional image provided above.

[0077] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this specification may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0078] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A method for determining the number of slice layers of a three-dimensional image, characterized in that: include: Determine the initial slice layer number N and the initial search range n, where both N and n are positive integers. Based on the initial slice layer number N and the initial search range n, determine the left neighboring slice layer number Nn and the right neighboring slice layer number N+n. Restored 3D images corresponding to the initial slice layer number N, the left neighboring slice layer number Nn, and the right neighboring slice layer number N+n are obtained respectively, and 2D plane images corresponding to the restored 3D images are collected from different angles. Based on the pixel connectivity or edge pixel density in the 2D plane images at different angles, characteristic values ​​used to characterize the quality of each restored 3D image are obtained; The number of slice layers corresponding to the optimal eigenvalue is used as the new initial number of slice layers, and the initial search range is narrowed; the eigenvalue acquisition operation is repeated according to the new initial number of slice layers and the reduced initial search range until the fluctuation of the eigenvalue is less than the set threshold, and the optimal number of slice layers is obtained; The characteristic values ​​include a connectivity characteristic value P1 for characterizing voxel connectivity in the three-dimensional image. The characteristic values ​​for characterizing the quality of each restored three-dimensional image are obtained based on the pixel connectivity or edge pixel density in the two-dimensional plane image at different angles, specifically including: Collect two-dimensional plane images of the restored three-dimensional image corresponding to the slice layer number i from different angles; Use the connected domain analysis method to obtain the number of connected pixel sets in each two-dimensional plane image; Take the weighted sum of the number of connected pixel sets to obtain the eigenvalue P1 corresponding to the slice layer number i: P1=(n 1 +n 2 +…+n J ) / J Among them, n 1 ~n J are the numbers of connected pixel sets in the J two-dimensional plane images obtained from J angles respectively. The above formula means that the quality of the restored three-dimensional image is characterized by the number of connected pixel sets at different angles: the larger the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the worse the restoration quality; the smaller the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the better the restoration quality.

2. The method for determining the number of slice layers of a three-dimensional image according to claim 1, wherein: The characteristic value includes an edge characteristic value P2 for characterizing the edge clarity of the three-dimensional image, specifically including: Collect two-dimensional plane images of the restored three-dimensional image corresponding to the slice layer number i from different angles; Obtain edge pixel density in each two-dimensional plane image based on an edge detection algorithm; The weighted sum of all edge pixel densities is used to obtain the eigenvalue P2 corresponding to the slice layer number i: P2=(m 1 +m 2 +…+m J ) / J Among them, m 1 ~m J are the edge pixel densities in the J two-dimensional plane images obtained from J angles respectively. The above formula means that the quality of the restored three-dimensional image is characterized by the edge pixel density at different angles: the greater the edge pixel density in each two-dimensional plane image, the clearer the edge in the restored three-dimensional image and the better the restoration quality; the smaller the edge pixel density in each two-dimensional plane image, the blurrier the edge in the restored three-dimensional image and the worse the restoration quality.

3. The method for determining the number of slice layers of a three-dimensional image according to claim 1, wherein: The restored three-dimensional image is obtained by scanning a shaped linear light source in a display material using a scanning galvanometer to form a continuous multi-layer screen, and projecting different sections of the three-dimensional model onto different screens in sequence through a digital micromirror display.

4. The method for determining the number of slice layers of a three-dimensional image according to claim 3, wherein: The wavelength of the linear light source is 1550 nm, and the curtain is a single-frequency excited curtain formed along the path of the linear light source when the linear light source passes through tellurite glass doped with trivalent erbium ions Er3+.

5. The method for determining the number of slice layers of a three-dimensional image according to claim 4, wherein: The digital micromirror display projects a pattern formed by irradiating a light source with a wavelength of 850nm onto tellurite glass doped with trivalent erbium ions Er3+, forming a dual-frequency excitation to generate a green pattern on the screen.

6. A device for determining the number of slice layers of a three-dimensional image, characterized in that: include: A search initialization module is used to determine the initial number of slice layers N and the initial search range n, where both N and n are positive integers. Based on the initial number of slice layers N and the initial search range n, the number of left neighboring slice layers Nn and the number of right neighboring slice layers N+n are determined. A quality evaluation module is used to obtain restored 3D images corresponding to the initial slice layer number N, the left neighboring slice layer number Nn, and the right neighboring slice layer number N+n, respectively, collect 2D plane images corresponding to the restored 3D images from different angles, and obtain characteristic values ​​used to characterize the quality of each restored 3D image based on the pixel connectivity or edge pixel density in the 2D plane images at different angles; A slice layer number determination module is configured to use the number of slice layers corresponding to the optimal eigenvalue as a new initial number of slice layers and narrow the initial search range; repeatedly perform the eigenvalue acquisition operation based on the new initial number of slice layers and the narrowed initial search range until the fluctuation of the eigenvalue is less than a set threshold, thereby obtaining the optimal number of slice layers; The characteristic value includes a connectivity characteristic value P1 for characterizing connectivity of voxels in a three-dimensional image. The quality evaluation module is specifically configured to: Collect two-dimensional plane images of the restored three-dimensional image corresponding to the slice layer number i from different angles; Use the connected domain analysis method to obtain the number of connected pixel sets in each two-dimensional plane image; Take the weighted sum of the number of connected pixel sets to obtain the eigenvalue P1 corresponding to the slice layer number i: P1=(n 1 +n 2 +…+n J ) / J Among them, n 1 ~n J are the numbers of connected pixel sets in the J two-dimensional plane images obtained from J angles respectively. The above formula means that the quality of the restored three-dimensional image is characterized by the number of connected pixel sets at different angles: the larger the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the worse the restoration quality; the smaller the number of connected pixel sets in each two-dimensional plane image, the more dispersed the voxels in the restored three-dimensional image, and the better the restoration quality.

7. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

Images