JP2019034487A - Image processing apparatus and image processing program - Google Patents

Abstract

To create a three-dimensional object with edges properly represented.SOLUTION: An edge is detected for voxels of data of a three-dimensional object in which gradation values are assigned to voxels representing a three-dimensional space, and a detection result of the edge for a certain voxel B2 is assigned as a detection result of the edge for voxel B1 aligned with the voxel B2 in a vertical direction with respect to a surface of the three-dimensional object.SELECTED DRAWING: Figure 18

Description

  The present invention relates to an image processing apparatus and an image processing program.

  A three-dimensional modeling apparatus that models a three-dimensional object by stacking cross-sectional shapes based on shape data of a three-dimensional model is known. For example, an optical modeling type three-dimensional printer that forms a three-dimensional shape by forming and stacking cross-sectional shapes by discharging ultraviolet curable ink on a stage is known. In such a three-dimensional modeling apparatus, it is possible to form a three-dimensional object reflecting not only the shape but also the color by ejecting color ink onto a portion that becomes the surface of the three-dimensional object.

  In Patent Document 1, cyan (C), magenta (M), yellow (Y), and white (W) toners are used to form a color toner layer after covering the entire surface of a transparent sheet with a white toner layer. Thus, a method of manufacturing a three-dimensional structure with improved surface color gradation and color tone is disclosed. In Patent Document 2, an error diffusion process is applied to a surface pixel aggregate corresponding to a surface portion of a modeled object in an apparatus that forms a three-dimensional modeled object by solidifying a three-dimensional modeled powder with a modeling liquid. A three-dimensional modeling data creation device that suppresses vertical stripes of color generated in a modeled object by creating modeling data is disclosed.

JP 2000-246804 A JP 2015-44299 A

  By the way, when modeling a three-dimensional object in a three-dimensional modeling apparatus, there is a desire to improve and clearly reproduce the blurring of edges such as characters and lines on the object surface. For a two-dimensional image, a method is disclosed in which an edge is emphasized by changing image processing applied between pixels in an edge portion and pixels that are not an edge portion. However, there is no known method for appropriately processing edges on the surface of a three-dimensional object.

  According to a first aspect of the present invention, there is provided an edge detection means for detecting an edge for each pixel in the data of a three-dimensional object to which a gradation value is assigned for each pixel representing a three-dimensional space, and a pixel by the edge detection means. And a detection result allocating means for allocating the edge detection result for the pixel as an edge detection result for a pixel perpendicular to the surface of the three-dimensional object from the pixel.

  The invention according to claim 2 is characterized in that the edge detection means detects an edge with respect to a pixel in which an application range of the edge detection filter is within a range of a pixel to which a gradation value is assigned. The image processing apparatus according to claim 1.

  According to a third aspect of the present invention, the detection result assigning means assigns an edge detection result to a pixel in which an application range of the edge detection filter does not fall within a pixel range to which a gradation value is assigned. The image processing apparatus according to claim 1, wherein:

  According to a fourth aspect of the present invention, there is provided gradation correction means for correcting a gradation value of a pixel using an edge detection result assigned by the detection result assignment means. An image processing apparatus according to any one of the preceding claims.

  According to a fifth aspect of the present invention, there is provided an edge detection means for detecting an edge for each pixel in the data of a three-dimensional object to which a gradation value is assigned for each pixel representing a three-dimensional space, and a pixel by the edge detection means. This is an image processing program that functions as a detection result assigning unit that assigns an edge detection result for a pixel as an edge detection result for a pixel perpendicular to the surface of the three-dimensional object from the pixel. .

  According to the invention described in claim 1 or 5, an appropriate edge detection result can be extended and assigned to a pixel to which the edge detection filter cannot be directly applied.

  According to the second aspect of the present invention, an edge can be appropriately detected by applying the edge detection filter to pixels within the application range of the edge detection filter.

  According to the third aspect of the present invention, the edge detection result for other pixels can be appropriately assigned to pixels that are not within the application range of the edge detection filter.

  According to the fourth aspect of the present invention, the gradation value of the pixel can be appropriately corrected based on the edge detection result appropriately set for each pixel.

It is a figure which shows the structure of the three-dimensional modeling apparatus in embodiment of this invention. It is a figure which shows the structure of the modeling part in embodiment of this invention. It is a figure which shows the structure of the ink head in embodiment of this invention. It is a figure which shows the structure of the control part in embodiment of this invention. It is a flowchart which shows the color signal process and modeling process in embodiment of this invention. It is a functional block diagram of the three-dimensional modeling apparatus in embodiment of this invention. It is a figure explaining the voxelization process in embodiment of this invention. It is a figure explaining the voxel used as the object of edge detection in an embodiment of the invention. It is a figure which shows the example of the edge detection filter in embodiment of this invention. It is a figure which shows another example of the edge detection filter in embodiment of this invention. It is a figure which shows another example of the edge detection filter in embodiment of this invention. It is a figure which shows another example of the edge detection filter in embodiment of this invention. It is a figure explaining the structure of the edge detection filter in embodiment of this invention. It is an exploded view of the edge detection filter in embodiment of this invention. It is an exploded view of the edge detection filter in embodiment of this invention. It is an exploded view of the edge detection filter in embodiment of this invention. It is an exploded view of the edge detection filter in embodiment of this invention. It is a figure for demonstrating the detection result allocation process of the edge in embodiment of this invention. It is a figure explaining the process which specifies the surface of the three-dimensional object in embodiment of invention.

  A three-dimensional modeling apparatus 100 according to an embodiment of the present invention includes a modeling unit 102 and a control unit 104 as shown in FIG. The control unit 104 generates modeling data considering the surface color from the three-dimensional data representing the three-dimensional object, and controls the modeling unit 102 based on the modeling data to model the three-dimensional object.

  1 shows a configuration in which the modeling unit 102 and the control unit 104 are directly connected, but the modeling unit 102 and the control unit 104 may be connected via an information communication network such as the Internet.

  As shown in FIG. 2, the modeling unit 102 includes a modeling table 10, a stage 12, an ink tank 14, an ink head 16, and a head driving unit 18. The modeling table 10 is a component that serves as a foundation for mechanically supporting the modeling unit 102. On the modeling table 10, a stage 12, an ink tank 14, an ink head 16, and a head driving unit 18 are arranged.

  The stage 12 is a component serving as a table for forming a modeled object. The upper surface of the stage 12 is preferably a shape suitable for forming a modeled object, for example, a flat surface. In the present embodiment, the vertical direction (upward direction) is described as the Z-axis direction for convenience. The stage 12 includes a stage driving mechanism. The stage 12 can be moved up and down along the Z-axis direction according to a control signal from the control unit 104.

  The ink tank 14 is a component for storing ink for constituting a three-dimensional object to be shaped. In the present embodiment, since the modeled object is a three-dimensional object having a color surface, the ink tank 14 includes an ink storage unit that stores ink of color components constituting the output color. In the present embodiment, the color components constituting the output color are cyan (C), magenta (M), and yellow (Y). However, it is not limited to this. The ink tank 14 also includes an ink storage unit that stores base color ink. The base color is the color of ink that forms an internal region other than the surface region of the three-dimensional object to be shaped. The base color is preferably a bright achromatic color. Specifically, the base color may be white (W) or transparent (T). In this embodiment, the base color is white (W).

  The ink head 16 is a component that discharges ink supplied from the ink tank 14. The ink head 16 includes a number of ink nozzles that can eject the color components constituting the output color and the base color ink. The ink head 16 is disposed on the stage 12.

  FIG. 3 shows a configuration example of the ink head 16. The ink head 16 includes a nozzle group 16A in which ink nozzles 16C, 16M, 16Y, and 16W corresponding to CMYW colors are arranged in the Y-axis direction. The ink head 16 has a nozzle array structure in which a plurality of nozzle groups 16A are arranged along the X-axis direction. It is preferable that the nozzle group 16 </ b> A is arranged as many as necessary to form a modeled object across the width of the stage 12 in the X-axis direction.

  Further, when the ink used in the modeling unit 102 is an ultraviolet curable type, the ultraviolet irradiation means 16 </ b> V may be provided in the ink head 16. As the ultraviolet irradiation means 16V, an ultraviolet lamp, an ultraviolet laser, or the like can be applied. The ultraviolet irradiation means 16V may be provided separately from the ink head 16.

  The head drive unit 18 is a component for moving the ink head 16 on the stage 12. The head drive unit 18 includes a rail 18a arranged along the Y-axis direction. Further, it may be configured to include a motor (not shown) for moving the ink head 16 along the rail 18a. The ink head 16 is supported by a rail 18a so as to be movable in the Y-axis direction, and is coupled to a motor via a driving member such as a ball screw. The head driving unit 18 can move the ink head 16 along the rail 18a, that is, along the Y-axis direction by rotating the motor in accordance with a control signal from the control unit 104. However, the configuration of the head drive unit 18 is not limited to this, and any configuration may be used as long as the ink head 16 can be moved relative to the stage 12 in order to form a three-dimensional object.

  The modeling unit 102 receives from the control unit 104 modeling data including position information of pixels (voxels: volume elements) constituting the three-dimensional object and color information of the voxels. The modeling unit 102 moves the stage 12 up and down to a position in the Z-axis direction corresponding to the position information received from the control unit 104, and moves the ink head 16 to a position in the Y-axis direction corresponding to the position information. Then, the ink of the color specified by the color information of the voxel is ejected from any one of the ink nozzles 16C, 16M, 16Y, and 16W in the X-axis direction corresponding to the position information. After the ink is ejected from the ink nozzles 16C, 16M, 16Y, and 16W, the ink can be cured by irradiating the ultraviolet rays from the ultraviolet irradiation means 16V.

  A three-dimensional object is represented by three-dimensional data including a group of voxels (volume elements) which are lattice units in the X-axis direction-Y-axis direction-Z-axis direction. The three-dimensional data has one layer for each voxel thickness along the X-axis-Y-axis plane. The modeling unit 102 ejects ink along the X-axis-Y-axis plane while moving the ink head 16 in the Y-axis direction. When modeling for one layer is completed, the stage 12 is moved downward along the Z-axis direction. Let me move on to the next layer. The modeling unit 102 models a three-dimensional object by repeating this.

  As shown in FIG. 4, the control unit 104 includes a processing unit 30, a storage unit 32, an input unit 34, an output unit 36, and a communication unit 38. The control unit 104 can be configured by a general computer. The processing unit 30 includes means for performing arithmetic processing such as a CPU. The processing unit 30 executes the image processing program stored in the storage unit 32 to realize the processing of images and color signals in the present embodiment. Thereby, the control unit 104 functions as an image processing apparatus. In addition, the control unit 104 executes a modeling process program stored in the storage unit 32 to output and control a signal that causes the modeling unit 102 to eject ink. The storage unit 32 includes storage means such as a semiconductor memory or a memory card. The storage unit 32 is connected to the processing unit 30 so as to be accessible, and stores an image processing program, a modeling processing program, and information necessary for the processing. The input unit 34 includes means for inputting information. The input unit 34 includes, for example, a keyboard, a touch panel, and a button that receive a command input from the user. The output unit 36 presents information regarding processing to the user. The output unit 36 can be, for example, a display that displays an image to the user. The communication unit 38 includes an interface for outputting information such as position information and color information to the modeling unit 102. Moreover, the communication part 38 may be provided with the interface for acquiring the three-dimensional data of the three-dimensional object used as modeling object from the outside via information communication networks, such as the internet.

  In the present embodiment, the control unit 104 is realized by executing a program in a general computer, but part or all of the control unit 104 may be realized by hardware such as an ASIC.

  Various programs such as a color signal processing program and a modeling processing program may be provided via the communication unit 38 or may be provided by a recording medium such as a CDROM.

  Hereinafter, with reference to the flowchart shown in FIG. 5, the modeling process of the three-dimensional object in this Embodiment is demonstrated. The control unit 104 executes the following processing by executing the image processing program and the modeling processing program. The three-dimensional modeling apparatus 100 functions as each unit shown in the functional block diagram of FIG. 6 by the following processing.

  In step S10, three-dimensional data acquisition processing is performed. The control unit 104 functions as the three-dimensional data acquisition unit 150 by the processing in this step. The processing unit 30 acquires three-dimensional data representing the outer shape of the three-dimensional object to be modeled via the input unit 34 or the communication unit 38. The acquired three-dimensional data is stored in the storage unit 32. The three-dimensional data includes three-dimensional object shape data designed by CAD or the like, or three-dimensional object shape data measured by a three-dimensional shape measuring instrument or the like.

  The three-dimensional data also includes color data (first color signal) of the surface of the three-dimensional object. That is, the three-dimensional data acquisition unit 150 includes a color signal acquisition unit. In the present embodiment, the three-dimensional data includes color data representing the surface color of a three-dimensional object in an RGB color space composed of red (R signal), green (G signal), and blue (B signal) color components. Is included.

  In step S12, voxelization processing is performed. By the processing in this step, the control unit 104 functions as the voxelization processing unit 152. The processing unit 30 reads the three-dimensional data acquired in step S10 from the storage unit 32, and converts the three-dimensional data into voxel data. The voxel data is data representing a three-dimensional object represented by three-dimensional data as a set of normal lattice units (voxels) in a three-dimensional space. In the voxel data, each voxel includes information indicating the position of the voxel, information indicating whether or not a three-dimensional object defined by the three-dimensional data is configured, and information indicating a distance from the object surface.

  The voxelization process is performed as follows, for example. Data indicating that the three-dimensional space is divided into voxels having a shape and size as a unit of division and the voxels included in the space covered by the surface of the three-dimensional object constitute a three-dimensional object (for example, “1”) is assigned, and data (for example, “0”) indicating that a three-dimensional object is not configured is assigned to voxels outside the space. More specifically, for example, as shown in FIG. 7, the three-dimensional object 200 is cut out with the width of the voxel along the Z-axis direction. Furthermore, the inside of the cross-sectional shape of the object in the cut out X-axis-Y-axis plane is divided into a voxel width grid to obtain a voxel B. Then, data indicating that a three-dimensional object is configured is assigned to each voxel B. Further, the shortest distance from the surface of the three-dimensional object to the center point of the voxel B is calculated and assigned to each voxel B. Such processing is repeated along the Z-axis direction of the three-dimensional object, and voxel processing is performed by stacking voxel layers. The generated voxel data is stored in the storage unit 32.

  In step S14, color information is assigned to voxels. The control unit 104 functions as the color information assigning unit 154 by the processing in this step. The processing unit 30 assigns the color data (first color signal) included in the three-dimensional data to each voxel of the voxel data obtained in step S12. Among the voxels to which data indicating that a three-dimensional object is configured is assigned, the color data (first color signal) of the surface position of the three-dimensional data is mapped to the voxel located on the surface of the three-dimensional object. And assign. Further, the color data of the voxel on the object surface closest to the voxel is assigned to the voxel located inside the three-dimensional object.

  Note that color data may be assigned only to voxels within the reference distance D1 from the surface of the three-dimensional object, and a base color may be assigned to deep voxels exceeding the distance D1. That is, with the distance D1 as a threshold value, a surface color is assigned to a voxel located on the surface side of the distance D1, and a base color is assigned to a voxel located on the object inside of the distance D1. Further, the color data value assigned according to the distance of the voxel from the object surface may be weighted. For example, weighting is performed so that the surface color value decreases as the distance from the surface of the object increases, and the base color value increases. In addition, the value of the color data to be assigned may be weighted according to the deviation of the voxel position from the normal direction with respect to the object surface. For example, the value of color data weighted by the cosine value of the angle shifted from the normal direction of the voxel may be assigned to the voxel. Furthermore, these methods may be combined.

  In step S <b> 16, color space conversion processing is performed for converting the color data (first color signal) acquired as the three-dimensional data into color data for representing the output color output to the modeling unit 102. By this processing, the control unit 104 functions as the color space conversion unit 156. The three-dimensional data acquired in step S10 includes color data (first color signal) composed of a first color space composed of a first color component (RGB in the present embodiment). The output color output to the modeling unit 102 is represented by color data including a second color space composed of a second color component (CMY in the present embodiment) that can be expressed by the ink used in the modeling unit 102. The In this step, color space conversion processing from the first color space to the second color space is performed for each voxel to which color data is assigned in step S14.

In the present embodiment, the output color is a CMY color composed of cyan (C ₁ signal), magenta (M ₁ signal), and yellow (Y ₁ signal) color components that can be expressed by the ink used in the modeling unit 102. It is represented by color data (color signal) expressed in space. However, the color components constituting the output color are not limited to these. For color space conversion from the RGB color space to the CMY color space, existing complementary color conversion processing or conversion processing using a 3D-LUT (three-dimensional lookup table) may be applied.

  In step S18, edge detection processing is performed. By this processing, the control unit 104 functions as the edge detection unit 158. The control unit 104 detects whether each voxel corresponds to an edge three-dimensionally.

  Specifically, each voxel constituting the object in the three-dimensional data is set as the target voxel in order, and it is determined whether the target voxel corresponds to the edge portion based on the relationship with the neighboring voxels in the vicinity of the target voxel. FIG. 8 is a diagram for explaining a collection 300 of voxels to be subjected to edge detection. The aggregate 300 is composed of an aggregate 302 composed of nine voxels along the X-axis-Y-axis plane around the target voxel A, an aggregate 304 obtained by rotating the aggregate 302 along the Y-axis, and an aggregate 302 Are combined with an assembly 306 that is rotated along the X axis.

  The control unit 104 applies edge detection processing to the target voxel A for each of the aggregates 302, 304, and 306. FIG. 9 shows an example of an edge detection filter (edge detector). In FIG. 9, the first edge detection filter, the second edge detection filter, and the third edge detection filter are applied to each of the aggregates 302, 304, and 306. The first edge detection filter performs summation of gradations of three voxels B (voxels hatched with left oblique lines) and three voxels C (right oblique hatching) around the target voxel A in the aggregate 302. It is determined that the target voxel A corresponds to the edge when the difference between the gradation of the applied voxel) and the sum of gradations is larger than the reference value. Similarly, the second edge detection filter and the third edge detection filter are the sum of the gray levels of the three voxels B (the voxels hatched with left oblique lines) around the target voxel A for the aggregates 304 and 306, respectively. And the three voxels C (voxels hatched with right oblique lines) are larger than the reference value, it is determined that the target voxel A corresponds to an edge.

  10 to 12 show other examples of the edge detection filter. As in FIG. 9, in the aggregates 302, 304, and 306, the sum of the gray levels of the three voxels B (the voxels with hatched left diagonal lines) around the target voxel A and the three voxels C (right diagonal lines) When the difference between the shaded voxel) and the sum of gradations is larger than the reference value, it is determined that the target voxel A corresponds to an edge.

Note that the gradation of each voxel can be the intensity of each voxel color signal (C ₁ signal, M ₁ signal, Y ₁ signal). In this case, it is determined whether each color of each voxel corresponds to an edge. The gradation of each voxel can be the intensity of the entire color signal of each voxel. The intensity of the entire color signal may be a value such as the sum or square root of each color signal (C ₁ signal, M ₁ signal, Y ₁ signal). In this case, it is determined whether the intensity of the entire color of each voxel corresponds to an edge.

  FIG. 13 shows another example of the edge detection filter. In this example, the edge detection process is applied to 27 voxels constituting the cube 310. The target voxel A is a voxel at the center of the cube 310. The cube 310 is formed by combining the uppermost layer aggregate 312, the intermediate layer aggregate 314, and the lowermost layer aggregate 316 including nine voxels along the X-axis / Y-axis plane with the target voxel A as the center. The

  The control unit 104 applies edge detection processing to the target voxel A by combining the aggregates 312, 314, and 316. FIG. 14 shows an example of an edge detection filter. The sum of the gradation of the voxel B located at the upper left corner of the cube 310 and the three voxels B adjacent to it (hatched with hatched left diagonal lines) and the voxel C located at the lower right corner and adjacent thereto It is determined that the target voxel A corresponds to an edge when the difference between the gradations of the three voxels C (the voxels hatched with right oblique lines) is larger than the reference value. 15 to 17 show edge detection filters obtained by rotating the edge detection filter shown in FIG. 14 by 90 °, 180 °, and 270 ° counterclockwise along the Z axis, respectively. Also in these edge detection filters, the sum of the gradations of four voxels B (voxels with hatched left diagonal lines) and four voxels C (voxels with hatched right diagonal lines) around the target voxel A When the difference from the sum of gradations is larger than the reference value, it is determined that the target voxel A corresponds to an edge.

  Note that the above edge detection filters may be used in combination. By using a plurality of types of edge detection filters in combination, erroneous detection of edges due to the influence of noise or the like can be suppressed.

  By the way, in the vicinity of the surface of the three-dimensional object, there are voxels on the surface of the three-dimensional object having gradation values (color information) and voxels inside the three-dimensional object, and objects having no gradation values (color information). A mix of external voxels. A voxel located outside the three-dimensional object, that is, a voxel that does not constitute a three-dimensional object does not have a gradation value. Therefore, on the surface of the three-dimensional object, there is a possibility that edge detection cannot be performed properly due to the influence of voxels outside the object. Therefore, the process of step S20 is applied to voxels that constitute a three-dimensional object and that cannot detect edges.

  In step S20, processing for assigning the edge detection result obtained by the edge detection means 158 as the detection result of the edge of another voxel is performed. With this processing, the control unit 104 functions as the detection result assignment unit 160.

  FIG. 18 is a diagram illustrating a structure 320 obtained by cutting out a part near the surface of a three-dimensional object. The structure 320 does not include the uppermost structure 322 outside the structure 320, the structure 324 including the voxels that configure the surface of the structure 320, the structure 326, and the voxels that configure the surface of the structure 320. It can be decomposed into structures 328 and 330 composed only of voxels constituting the inside. In FIG. 18, voxels constituting the three-dimensional object are indicated by solid lines, and voxels not constituting the three-dimensional object are indicated by broken lines.

  Here, the application range of the edge detection filter shown in FIGS. When an edge detection filter is applied to the voxel B2 included in the structure 326 constituting the structure 320, the application range of the edge detection filter is within the range of voxels to which gradation values are assigned. Therefore, the edge can be detected appropriately. On the other hand, when an edge detection filter is applied to the structure 324 and the voxel B1 included in the structure 326 constituting the surface of the structure 320, the edge detection filter is within the range of voxels to which the gradation value is assigned. Since the application range is within the range, the edge cannot be detected properly.

  Therefore, the detection result assignment unit 160 performs processing for assigning the edge detection result of the voxel B2 to which the edge detection filter can be applied to the voxel B1 to which the edge detection filter is not applicable. Specifically, the edge detection result of the voxel B2 to which the edge detection filter located in the vertical direction of the surface of the three-dimensional object can be applied is assigned to the voxel B1 to which the edge detection filter is not applicable.

  For example, the voxel B1 positioned at the center of the structure 326 is positioned closest to the voxel B1 along the surface normal direction (Z-axis direction) and is positioned at the center of the structure 328 to which the edge detection filter can be applied. The value of voxel B2 to be expanded is assigned. Further, for example, the voxel B1 located at the center of the structure 324 is also located closest to the voxel B1 along the surface normal direction (Z-axis direction), and the center of the structure 328 to which the edge detection filter can be applied. The value of voxel B2 located in is expanded and assigned.

  As described above, the voxel edge detection result to which the edge detection filter can be applied is expanded and applied to the voxel to which the edge detection filter cannot be directly applied in the three-dimensional object. Thereby, the presence / absence of an edge is also set for voxels to which the edge detection filter cannot be directly applied.

  Here, in a three-dimensional object, the distribution of gradation values in the surface of the surface is often handed over to the distribution of gradation values in the surface within the three-dimensional object. That is, it is highly possible that the in-plane distribution of gradation values of each layer of voxels constituting the colored region of the three-dimensional object is inherited from the object surface toward the inside. For example, the distribution of gradation values in the voxels on the outermost surface of a three-dimensional object is often close to the distribution of gradation values in the voxels of the second layer. Therefore, it is highly possible that the edge detection results are similar between voxels positioned in the direction perpendicular to the surface of the three-dimensional object. Therefore, in this embodiment, by assigning the detection result of the edge of the voxel positioned in the vertical direction of the surface to the voxel to which the edge detection filter cannot be applied, the more appropriate detection result of the edge is assigned.

  The surface of the three-dimensional object can be determined by the following method. A voxel constituting a cube of a predetermined size is selected, and a surface having the most voxels that do not constitute an object is determined as a surface in each of the six faces of the cube. As shown in FIG. 19, when a cube 332 composed of 27 voxels is selected, the top surface of the cube 332 that has the most voxels that do not constitute an object among the front surface, back surface, top surface, bottom surface, right side surface, and left side surface of the cube 332 Judged to be surface. Therefore, the surface normal direction of the cube 332 is set to a direction perpendicular to the upper surface, that is, the Z-axis direction.

  In the present embodiment, edge detection processing for voxel gradation has been described. However, the target of edge detection is not limited to gradation. Edge detection processing can be similarly applied to other attribute values of voxels.

  In step S22, gradation correction processing is performed on the voxel located at the edge. By the processing in this step, the control unit 104 functions as the gradation correction unit 162. The processing in the gradation correction unit 162 is not particularly limited. For example, if a three-dimensional edge enhancement filter is applied to each voxel determined to be an edge in step S18 and step S20. Good.

In step S24, color data including a second color component signal representing the second color component from the color data converted into the second color space and a base color signal corresponding to the magnitude of the first color signal (second color signal). ) Is generated. By this processing, the control unit 104 functions as the second color signal generation unit 164. In the present embodiment, the base color is white (W) as described above. That is, the processing unit 30 converts the second color component signal representing the second color component and the base color signal from the color data (C ₁ M ₁ Y ₁ signal) converted into the second color space and applied with the edge correction process. Is converted to color data (second color signal: C ₂ M ₂ Y ₂ W ₂ signal).

First, the processing unit 30 generates a W ₂ signal from the C ₁ M ₁ Y ₁ signal to which the edge correction process is applied. The W ₂ signal can be obtained by Expression (1). Here, C ₁ , M ₁ , Y ₁ and W ₂ indicate the magnitudes of the C ₁ signal, the M ₁ signal, the Y ₁ signal and the W ₂ signal, respectively, and are real numbers from 0 to 1.
(Equation 1)
W ₂ = (1-C ₁ ) (1-M ₁ ) (1-Y ₁ ) (1)

As shown in Equation (1), the W ₂ signal has a value corresponding to the magnitudes of the C ₁ signal, the M ₁ signal, and the Y ₁ signal. Here, the C ₁ signal, the M ₁ signal, and the Y ₁ signal are obtained according to the magnitudes of the first color signals (R signal, G signal, and B signal). Thus, W ₂ signal has a value corresponding to the magnitude of the first color signal.

Further, the processing unit 30, such that the color represented by the first color signal in a second color signal containing W ₂ signal is maintained, the color data of the second color component in accordance with the W ₂ signal (C ₁ Modified CMY data (C ₂ signal, M ₂ signal, Y ₂ signal) is generated by modifying the signal, M ₁ signal, Y ₁ signal). By this processing, the processing unit 30 functions as the second color signal correcting unit 166.

The corrected CMY data (C ₂ signal, M ₂ signal, Y ₂ signal) can be obtained by Equation (2).
(Equation 2)
_{C 2 = C 1 (1-} M 1) (1-Y 1) + C 1 M 1 (1-Y 1) / 2 + C 1 (1-M 1) Y 1/2 + C 1 M 1 Y 1 / 3
_{M 2 = (1-C 1} ) M 1 (1-Y 1) + C 1 M 1 (1-Y 1) / 2 + (1-C 1) M 1 Y 1/2 + C 1 M 1 Y 1 / 3
_{Y 2 = (1-C 1} ) (1-M 1) Y 1 + C 1 (1-M 1) Y 1/2 + (1-C 1) M 1 Y 1/2 + C 1 M 1 Y 1 / 3 (2)

In step S <b> 26, the modeling unit 102 performs a process for selecting the color of ink ejected to each voxel. By the processing in this step, the control unit 104 functions as the ink color selection unit 168. In the modeling unit 102 in the present embodiment, only one ink of cyan (C), magenta (M), yellow (Y), and base color (W) is ejected from the ink head 16 to one voxel. Can do. The processing unit 30 performs a process of selecting an ink color for each voxel based on the corrected CMY signal. The ink selection process is not particularly limited, but the color selection rate selected from cyan (C), magenta (M), yellow (Y), and base color (W) is C ₃ 'signal, M It is preferable to select a color so as to correspond to the ratio of the magnitude of each value of the ₃ ′ signal, the Y ₃ ′ signal, and the W ₃ ′ signal.

For example, the reference value a that caused by sequence generating means, C 3 _'signal> select cyan (C) in the case of a, (C _3' magenta in the case of the signal + M 3 _'signal)> a ( M) is selected, and when (C ₃ ′ signal + M ₃ ′ signal + Y ₃ ′ signal)> a, yellow (Y) is selected, and in other cases, the base color (W) is selected. Further, an ordered dither method may be applied. That is, a sequence (dither matrix) is applied as the reference value a for each voxel, cyan (C) is selected when C ₃ ′ signal> a, and when (C ₃ ′ signal + M ₃ ′ signal)> a. A process of selecting magenta (M), selecting yellow (Y) when (C ₃ ′ signal + M ₃ ′ signal + Y ₃ ′ signal)> a, and selecting a base color (W) otherwise. To do. Further, an error diffusion method (Error Diffusion) may be applied. In this method, a fixed value is used as the reference value a, but an error that occurs when an ink color is selected for a certain voxel is diffused in the selection process of surrounding voxels. Furthermore, ink may be selected for each voxel by the method disclosed in Japanese Patent Laid-Open No. 2006-295899.

  Further, a white noise method, a blue noise mask method, or the like may be applied. The white noise method is a method in which white noise (uniform random number) is applied as the reference value a for each voxel. In the blue noise mask method, the size of the dither matrix is increased so that the spatial frequency distribution of the dither matrix applied to the reference value a is equal to or higher than the spatial frequency at which the human visual sensitivity is low.

  In step S <b> 28, slicing processing for cutting out each layer to be modeled in the modeling unit 102 is performed. By this processing, the control unit 104 functions as the slice processing unit 170. Similar to the slicing in the Z-axis direction in the voxelization process in step S12, the processing unit 30 slices the voxels constituting the three-dimensional object along the Z-axis direction and cuts out one layer for each voxel. That is, the ink head 16 of the modeling unit 102 extracts a layer that can be ejected in one scan. For each layer, the ink color selected for each voxel included in the layer is read from the storage unit 32 and is used as modeling data to be output to the modeling unit 102. The modeling data includes information indicating the position of each voxel and information on the color of the selected ink.

  In step S <b> 30, an ink ejection process is performed in the modeling unit 102. By the processing in this step, the modeling unit 102 functions as the ink ejection unit 172. The processing unit 30 ejects ink of a selected color for each voxel for each layer constituting the three-dimensional object. Each layer is stacked by repeating the ink ejection process from the lowermost layer to the uppermost layer in order along the Z-axis direction. Thereby, the three-dimensional object based on the input three-dimensional data is modeled.

  In the three-dimensional modeling apparatus 100 according to the present embodiment, the modeling unit 102 that ejects only one color ink per voxel is used to improve the granularity of the surface color and appropriately set the intermediate color of the ink that can be ejected. The three-dimensional object expressed in In addition, the three-dimensional object is formed after appropriately detecting the edge in the three-dimensional object and performing processing for enhancing the edge.

  10 modeling stand, 12 stage, 14 ink tank, 16 ink head, 16A nozzle group, 16C, 16M, 16Y, 16W ink nozzle, 16V ultraviolet irradiation means, 18 head drive unit, 18a rail, 30 processing unit, 32 storage unit, 34 input unit, 36 output unit, 38 communication unit, 100 three-dimensional modeling apparatus, 102 modeling unit, 104 control unit, 150 three-dimensional data acquisition unit, 152 voxelization processing unit, 154 color information allocation unit, 156 color space conversion unit 158 Edge detection means 160 Detection result assignment means 162 Gradation correction means 164 Color signal generation means 166 Color signal correction means 168 Ink color selection means 170 Slice processing means 172 Ink ejection means 200 Three-dimensional object , 302, 304, 306 aggregate, 310 standing Cuboid, 312, 314, 316 aggregate, 320-330 structure, 332 cube.

Claims (5)

Edge detection means for detecting an edge for each pixel of data of a three-dimensional object in which gradation values are assigned to each pixel representing a three-dimensional space;
A detection result assigning means for assigning an edge detection result for the pixel by the edge detection means as an edge detection result for a pixel perpendicular to the surface of the three-dimensional object from the pixel;
An image processing apparatus comprising:   The image processing according to claim 1, wherein the edge detection unit detects an edge of a pixel in which an application range of the edge detection filter is within a range of pixels to which gradation values are assigned. apparatus.   3. The detection result assigning means assigns an edge detection result to a pixel in which an application range of an edge detection filter does not fall within a pixel range to which a gradation value is assigned. An image processing apparatus according to 1.   The image processing according to claim 1, further comprising a gradation correction unit that corrects a gradation value of a pixel using an edge detection result assigned by the detection result assignment unit. apparatus. Edge detection means for detecting an edge for each pixel of data of a three-dimensional object in which gradation values are assigned to each pixel representing a three-dimensional space;
A detection result assigning means for assigning an edge detection result for the pixel by the edge detection means as an edge detection result for a pixel perpendicular to the surface of the three-dimensional object from the pixel;
An image processing program characterized by being made to function.

Images