3D object image scanner

Abstract

Provided is a three-dimensional object image scanner that can create a full circumference image that is suitable for the observation of the entire circumference of a three-dimensional object and that has less distortions.　This three-dimensional object image scanner comprises: a rotational table; a scanner head having a telecentric lens system; a shifting means that shifts the scanner head; and a control device that processes the image data of a subject. The control device acquires multiple pieces of image data for which the rotation angles are different and which are sequentially outputted from the scanner head, while rotating the subject about the center axis with the scanner head located in such a manner that the central axis is in front of the telecentric lens system; detects, from among the multiple pieces of image data, pixel data for which the subject is focused; integrates the detected pixel data, thereby creating a rectangular image representative of the outward appearance of a portion of the subject that is desirably to be imaged; and converts the rectangular image to an annular fan image based on the upper surface dimension, bottom surface dimension and height dimension of the portion of the subject that is desirably to be imaged.

Description

Technical Field

[0001] The present invention relates to a three-dimensional object image scanner that optically scans the outer surface of a three-dimensional object to create an appearance image thereof.

Background Art

[0002] Patent Document 1 discloses a method for forming an orthographic projection image of a three-dimensional object. The method relatively scans a three-dimensional object with a camera equipped with a telecentric optical system, and among the image light obtained in this scanning, a predetermined light ray is selected and stored as image information. By processing this stored information and extracting the image light within the depth of field of the optical system, an orthographic projection image of the three-dimensional object is formed.

[0003] Patent Document 2 discloses an image scanner for obtaining an orthographic projection image of a three-dimensional object. The apparatus includes a telecentric imaging system disposed between an image sensor and a target plane region located at a position away from the image sensor in the observation direction, and forms an image of the target plane region on the image sensor. The target plane region is moved in a three-dimensional direction within the subject space where the subject is disposed, and while regulating the direction of the illumination light so that the distribution range of the incident angle of the illumination light on the target plane region is constant at every point of the target plane region, the target plane region is irradiated with the illumination light. Image data sequentially output from the image sensor while the target plane region is moving in the subject space is input, pixel data in focus on the subject is detected from the input image data, and the detected pixel data is collected to synthesize an orthographic projection image of the subject.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

[0005] In fields such as archaeology, history, and art history, artifacts such as vessels, jars, bottles, and bowls are often studied, and these artifacts frequently have patterns or decorations covering their entire 360-degree surface. Researchers have a need for 360-degree images suitable for observing the patterns and decorations on such artifacts.

[0006] However, the shape of these objects is rarely a simple cylinder with a constant diameter at all height positions; rather, they are complex three-dimensional shapes with varying diameters at different height positions. Therefore, when optically scanning such objects using a camera (image sensor) equipped with a telecentric optical system to generate an orthographic projection image, the shapes of patterns and decorations become distorted in the resulting orthographic projection image, making them inconvenient to observe.

[0007] The object of the present invention is to provide a stereoscopic image scanner capable of creating a low-distortion, full-circumference image suitable for observing the entire outer circumference of a three-dimensional object. Other objects of the present invention will become apparent from the following description. [Means for solving the problem]

[0008] A stereoscopic object image scanner according to one embodiment includes a rotary table capable of rotating a subject around a predetermined central axis, a scanner head having a telecentric lens system and acquiring an image of the subject space in front of the telecentric lens system, a moving means for moving the scanner head relative to the subject space, and a control device that controls the rotary table, the scanner head, and the moving means and processes the image data of the subject.

[0009] The control device is With the scanner head positioned such that the central axis is located in front of the telecentric lens system, the subject is rotated around the central axis while acquiring multiple image data with different rotation angles that are sequentially output from the scanner head. From the plurality of image data, pixel data that is in focus on the subject is detected, and the detected pixel data is integrated to create a rectangular image that represents the appearance of the part of the subject that is to be photographed. The rectangular image is converted into an annular sector image based on the top, bottom, and height dimensions of the part of the subject to be photographed.

[0010] The control device may allow the user to set the portion of the subject to be photographed, as well as the top surface dimensions, bottom surface dimensions, and height dimensions.

[0011] The annular sector image may be configured such that the dimensional ratio between the dimension of its outer arc, the dimension of its inner arc, and the dimension of its radial width is equal to the dimensional ratio between the top surface dimension, the bottom surface dimension, and the height dimension.

[0012] The control device may output the rectangular image and the annular sector image.

[0013] The rotating table may be movable along the shooting direction of the scanner head.

[0014] One embodiment of a three-dimensional object image scanner can create a low-distortion, full-circumference image suitable for observing the entire outer circumference of a three-dimensional object. [Brief explanation of the drawing]

[0015] [Figure 1] This is the overall configuration of a 3D object image scanner according to one embodiment of the present invention. [Figure 2] Figure 1 is a flowchart of the scanning process performed by the 3D object image scanner. [Figure 3] This is a diagram showing the screen for setting the scan conditions for the scanning process. [Figure 4] This is an explanatory diagram illustrating the operation of scanning an object with a scanner head during rotational scanning. [Figure 5] This is an explanatory diagram of image processing in rotational scanning. [Figure 6] (a) Orthographic projection image, (b) rectangular image (orthographic projection image), and (c) annular sector image when an actual clay vessel is XY-scanned by a three-dimensional object image scanner. [Figure 7] A block diagram showing an example of a system and method for creating a three-dimensional view of a textured subject using a three-dimensional object image scanner. [Figure 8A] A diagram showing an example of the data structure of a rectangular image dataset output from a three-dimensional object image scanner. [Figure 8B] A diagram explaining the meaning of the example of the rectangular image dataset shown in Fig. 8A. [Figure 8C] A diagram showing three-dimensional cylindrical data represented by the example of the rectangular image dataset shown in Fig. 8A. [Figure 9] A diagram showing the flow of the process performed by the 3DCG device in the system example shown in Fig. 7.

MODE FOR CARRYING OUT THE INVENTION

[0016] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the embodiments described below do not limit the invention according to the claims, and not all of the elements and combinations thereof described in the embodiments are essential for the solution means of the invention.

[0017] Fig. 1 shows the overall configuration of a three-dimensional object image scanner 1 according to an embodiment of the present invention.

[0018] As shown in Fig. 1, the three-dimensional object image scanner 1 has a platform 2 installed on the floor. In the following description, for the convenience of explaining the three-dimensional positional relationship, an orthogonal three-dimensional coordinate system having X, Y, and Z axes as shown in Fig. 1 is defined. Here, the Y-axis is the vertical direction (height direction), and the X-axis and Z-axis are horizontal directions.

[0019] A columnar Z-track 3 is laid parallel to the Z-axis on platform 2. A rotary table 4 is mounted on the Z-track 3. The rotary table 4 has a rotary drive unit 4A and a table 4B. The rotary drive unit 4A incorporates an electric self-propelled device (not shown) and can reciprocate in a direction parallel to the Z-axis (lateral direction) within the length of the Z-track 3. The table 4B is fixed to a rotating shaft (not shown) of the rotary drive unit 4A and is configured to rotate together with this rotating shaft about a central axis 31. The space above the table 4B where the subject S is placed is called the subject space A. In this embodiment, the subject S is a three-dimensional object with a roughly circular horizontal cross-section, such as pottery, a jar, or a flower vase.

[0020] A columnar X-track 5 is laid parallel to the X-axis on platform 2. A columnar Y-track 6 is mounted on the X-track 5 parallel to the Y-axis. A columnar carriage 7 is mounted on the Y-track 6 parallel to the Z-axis. The Y-track 6 incorporates an electric self-propelled device (not shown) and can reciprocate within the length of the X-track 5 in a direction parallel to the X-axis (depth direction). The carriage 7 also incorporates an electric self-propelled device (not shown) and can reciprocate within the length of the Y-track 6 in a direction parallel to the Y-axis (height direction), as well as in a direction parallel to the Z-axis (lateral direction) within its own length. As a result, the carriage 7 can move three-dimensionally along the X-axis, Y-axis, and Z-axis.

[0021] A scanner head 8 and an illumination device (not shown) are fixed to the carriage 7. The scanner head 8 is an optical device for three-dimensionally scanning a subject S placed in the subject space A, and has a telecentric imaging unit, a variable aperture unit, and an image sensor unit arranged parallel to the Z axis. The scanner head 8 is capable of capturing visible light images and infrared light images.

[0022] The image sensor unit incorporates a linear image sensor arranged parallel to the Y-axis. The linear image sensor includes, for example, an array of photoelectric conversion elements with 3000 or more pixels arranged in a straight line parallel to the Y-axis at a pitch of 8 μm, and an A / D conversion device that converts the analog voltage signals output from the photoelectric conversion element array into digital pixel data.

[0023] The telecentric imaging unit incorporates a telecentric lens system and is positioned so that its optical axis is parallel to the Z-axis and faces the positive Z-axis direction (shooting direction). The telecentric imaging unit forms an image on the linear image sensor in the image sensor unit of a linear region (hereinafter referred to as the subject area R) of a predetermined length parallel to the Y-axis, located at a predetermined distance (working distance) from its front end along the optical axis in the shooting direction. In other words, the scanner head 8 is in focus on the subject area R. The subject area R is, in detail, defined by the length of the objective lens in the Y-axis direction, the working distance, and the depth of field. In this embodiment, the working distance of the scanner head 8 is 288 mm, and the depth of field is 10 mm.

[0024] The variable aperture unit is located within the telecentric imaging unit and incorporates a variable aperture diaphragm (hereinafter referred to as the variable aperture). The variable aperture unit is used to adjust the amount of light input to the linear image sensor and to adjust the depth of field.

[0025] The scanner head 8 described above moves three-dimensionally in the X, Y, and Z directions by the movement of the carriage 7, and the subject area R moves along with it. By using a telecentric lens system as the imaging optical system, it becomes possible to obtain an orthographic projection image of the subject S. The 3D object image scanner 1 has an XY scanning function and a rotation scanning function. The XY scanning function is a function that obtains a planar image of the XY plane of the subject S by moving the scanner head 8 along the XY plane while the subject S is not rotated by the rotary table 4. The rotation scanning function is a function that obtains a rectangular image and an annular sector image (unfolded image) of the entire outer surface of the subject S by rotating the subject S by the rotary table 4 and moving the scanner head 8 along the YZ plane.

[0026] A control device 9, which incorporates a computer, power supply circuits, and other electrical and electronic circuits, is installed on platform 2. The control device 9 is electrically connected to the rotary drive unit 4A, the X-axis 5, the Y-axis 6, the electric self-propelled device in the carriage 7, the linear image sensor in the scanner head 8, and lighting devices (not shown) via signal cables and power cables. The control device 9 has the function of rotating the table 4B by driving and controlling the rotary drive unit 4A. The control device 9 has the function of controlling the three-dimensional movement and position of the carriage 7 (and consequently controlling the scanning of the subject S by the scanner head 8) by driving and controlling the X-axis 5, the Y-axis 6, and the electric self-propelled device in the carriage 7. Furthermore, while the scanner head 8 is scanning the subject S, the control device 9 has the function of inputting image data sequentially output from the linear image sensor in the scanner head 8 and processing that image data to synthesize orthographic projection images of the subject (rectangular image and annular sector image (unfolded image)).

[0027] The 3D object image scanner 1 includes an input device, such as a keyboard or mouse, operated by the user to perform the scanning process described later; an output device, such as a display, to show the data; and an external output terminal (not shown) for inputting and outputting data to and from these devices. For example, the control device 9 has the aforementioned external output terminal 60, which is connected to a terminal 61, such as a personal computer used by the user. The terminal 61 may execute a computer program for using the 3D object image scanner 1 and function as the aforementioned input and output device. In this case, the control device 9 receives information from the terminal 61 for scanning and outputs the results of the scanning process to the terminal 61.

[0028] The movable distance (stroke) of the scanner head 8 is, for example, 520 mm in the X-axis direction, 750 mm in the Y-axis direction, and 260 mm in the Z-axis direction. Therefore, the effective capture range in XY scanning is X: 520 × Y: 750 × Z: 260 [mm]. In other words, in XY scanning, the maximum size of an object S whose entire image can be scanned is a diameter of 520 mm and a height of 750 mm.

[0029] The movable distance (stroke) of the rotary table 4 in the Z-axis direction is 100 mm. Therefore, the effective capture range in rotational scanning is φ:700 × Y:750 × Z:260 [mm]. In other words, the maximum diameter of a subject S whose entire image can be scanned in rotational scanning is 700 mm. Thus, because the rotary table 4 is movable in the Z-axis direction, it is possible to photograph subjects S with larger diameters. The rotation direction of the rotary table 4 (i.e., the rotation direction of the subject S) may be a specific single direction. This specific single direction is, for example, clockwise in a top-down plan view, as indicated by arrow 62 in Figure 1. Rotating the subject S clockwise means that the scanner head 8 scans the surface of the subject S from left to right. This is convenient for creating an orthographic projection image of the subject S using a common image format. This is also convenient when performing texture mapping of a three-dimensional model of the subject S using the orthographic projection image of the subject S, as will be described later.

[0030] Next, the scanning process performed by the 3D object image scanner 1 will be explained based on Figures 2 to 5.

[0031] Figure 2 is a flowchart of the scanning process performed by the 3D object image scanner 1, including terminal 61. Figure 3 shows the scan condition setting screen 10 for the scanning process, which is displayed on the display screen of terminal 61.

[0032] First, a preview image of the subject S placed on table 4B is obtained (step 100). The acquisition of the preview image is performed by the user clicking the preview button 11 on the scan condition setting screen 10 (Figure 3) displayed on an output means (not shown).

[0033] As shown in Figure 3, the scan condition setting screen 10 includes a preview button 11, a scan button 12, a scan setting unit 20, a visible light image display unit 30, and an infrared light image display unit 40.

[0034] By clicking the preview button 11, the control device 9 performs an XY scan and displays the visible light image S1 and infrared light image S2 of the subject S on the visible light image display unit 30 and the infrared light image display unit 40. Table 4B is also displayed on the visible light image display unit 30 and the infrared light image display unit 40 (step 102).

[0035] The user sets the start position 21, capture height 22, bottom position 23, pottery height 24, φ bottom surface 25, φ top surface 26, Z-axis range 27, scan mode 28, and layer setting 29 in the scan setting section 20 of the scan condition setting screen 10 on the display screen of terminal 61, and then clicks the scan button 12 (step 104 in Figure 2). The start position 21 is the height (Y-axis) position where image acquisition begins. The capture height 22 is the height (distance in the Y-axis direction) from the height position where image acquisition begins to the height position where it ends. The bottom position 23 is the height position of the bottom surface of the pottery (subject S). The pottery height 24 is the height (distance in the Y-axis direction) from the bottom surface to the top surface of the pottery (subject S). The settings for the starting position 21, capture height 22, base position 23, and pottery height 24 define which range in the height (Y-axis) direction of the subject S is to be photographed, and which range in the height direction of the output image it lies within. φ base 25 is the diameter of the base of the pottery (subject S). φ top 26 is the diameter of the top of the pottery (subject S). The Z-axis range 27 is the displacement range in the Z-axis direction of the subject area R when the scanner head 8 acquires the image (starting Z-axis position and ending Z-axis position (rotation center axis 31 is the origin on the Z-axis)). In the illustrated example, the starting position 21 is 50.0 mm, the capture height 22 is 450 mm, the base position 23 is 75 mm, the pottery height 24 is 300 mm, φ base 25 is 200 mm, and φ top 26 is 300 mm. The Z-axis range 27 is set to start at 150 mm and end at 100 mm. In other words, the scanner head 8 scans the Z-axis range of the subject S from 150-100 mm from the central axis 31 in the Z-axis direction.

[0036] These settings allow lines 32 and 33 representing the central axis 31, starting position 21, and capture height 22, and lines 34 representing the bottom surface position 23, pottery height 24, φ bottom surface 25, and φ top surface 26 to be displayed on the visible light image display section 30 and infrared light image display section 40 on the display screen. The visible light image display section 30 and infrared light image display section 40 are each divided into multiple bands B1 to B5 based on the length (effective length) in the Y-axis direction of the subject area R. Boundary lines K are shown at the boundaries between adjacent bands B1 to B5. A line 33 representing the capture height 22 is shown at the boundary between band B3 and band B4. In the illustrated example, the length in the Y-axis direction of bands B1 to B5 (length in the Y-axis direction of the subject area R) is 150 mm, and the length in the Y-axis direction of the visible light image display section 30 and infrared light image display section 40 is 750 mm.

[0037] In scan mode 28, the user sets the diameter of the subject S, the capture resolution [ppi], and the exposure time [ms]. In the illustrated example, the diameter of the subject S is φ300, the capture resolution is 400ppi, and the exposure time is 2.5ms. The rotation speed of table 4B is determined by setting these parameters.

[0038] Layer setting 29 sets the number of layers or the layer spacing. Layer setting 29 divides the 150-100 range of the Z-axis range 27 into multiple layers with different Z coordinates (layers L1 to L6 in Figure 4). The thickness of each layer is less than or equal to the depth of field. In the illustrated example, the depth of field is 10 mm and the Z-axis range 27 is 50 mm, so the number of layers is "6" according to the formula "number of layers = Z-axis range / depth of field + 1". Therefore, as shown in Figure 4, the 150-100 range Q of the Z-axis range 27 is divided into layers L1 to L6.

[0039] Figure 4 is an explanatory diagram illustrating the operation of scanning the subject S with the scanner head 8 during rotational scanning.

[0040] The rotational scanning process will be explained based on Figures 2 and 4. Rotational scanning is performed with the scanner head 8 positioned relative to the subject S such that the central axis 31 is directly in front of the scanner head 8. In rotational scanning, an unprocessed layer is selected (step 106). The control device 9 selects layer L1 and then selects band B1, which is an unprocessed band in layer L1 (step 108). The control device 9 moves the scanner head 8 to a position suitable for band B1 and layer L1 (step 110). In this state, the control device 9 rotates the subject S around the central axis 31 by the rotary table 4, for example clockwise in a plan view, for example, one or more rotations (more than 360 degrees), and the scanner head 8 captures images of the rotating subject S around its entire circumference (step 112). Multiple image data with different rotation angles are sequentially output from the scanner head 8 and stored in the control device 9.

[0041] The control device 9 determines whether imaging of all bands has been performed on the selected layer L1 (step 114). If it is determined that imaging of all bands has not been performed (step 114: No), the control device 9 returns to step 108, selects the unprocessed bands, and repeats the processes in steps 110 and 112. If it is determined that imaging of all bands has been performed (step 114: Yes), the control device 9 determines whether imaging of all layers has been performed (step 116). If it is determined that imaging of all layers has been performed (step 116: No), the control device 9 returns to step 106, selects the unprocessed layers, and repeats the processes in steps 108-114. If it is determined that imaging of all layers has been performed (step 116: Yes), the control device 9 proceeds to step 118.

[0042] The control device 9 detects and integrates in-focus pixel data that is in focus on the outer surface of the subject S from the stored image data of all layers L1 to L6 and all bands B1 to B3 to form a rectangular image 50 (see Figure 5) of the entire outer surface of the subject S (step 118). This rectangular image corresponds to the image of the entire outer surface of the subject S (orthographic projection image) obtained by Lambert equal-area cylindrical projection. The control device 9 converts the rectangular image 50 of the entire outer surface of the subject S into an annular sector image 51 (see Figure 5) (step 120). The control device 9 stores the rectangular image 50 and the annular sector image 51 and outputs them to an output means (not shown) (step 122).

[0043] Figure 5 is an explanatory diagram of image processing in rotational scanning.

[0044] As shown in Figure 5, a rectangular image 50 of the entire outer surface of the subject S is created based on the focus pixel data obtained by processing steps 106 to 116 of Figure 3 for the subject S. If the dimensions of the subject S are the diameter of the top surface α1, the diameter of the bottom surface β1, and the height γ1, then the ratio of the length of the long side α2 of the rectangular image 50 to the length of the short side γ2 of the rectangular image 50 (dimensional ratio) is equal to the ratio of the circumference of the top surface of the subject S πα1 to the height γ1 of the subject S (dimensional ratio). Therefore, in the rectangular image 50, the long side corresponding to the bottom surface of the subject S is the same length as the long side on the top surface of the subject S. As a result, for example, if the letters A to E are written on the side of the subject S in Figure 5 from top to bottom, the letters A to E in the rectangular image 50 will be distorted, with the lower letters being wider.

[0045] In step 120, the rectangular image 50 is converted into an annular sector image 51. In the annular sector image 51, the ratio (dimensional ratio) of the length α2 of the outer arc, the length β2 of the inner arc, and the radial width γ2 is equal to the ratio (dimensional ratio) of the circumference πα1 of the top surface of the subject S, the circumference πβ1 of the bottom surface of the subject S, and the height γ1 of the subject S. This provides an annular sector image 51 as seen when the subject S is observed from the side. As a result, in the annular sector image 51, the widths of the letters A to E are approximately the same, and the distortion of the letters A to E is corrected with less distortion. The radial width γ2 may correspond to the straight-line distance γ3 connecting the top and bottom surfaces of the subject S, or, if the angle of the line segment connecting the top and bottom surfaces with respect to the vertical direction is 45° or more, it may correspond to the horizontal width γ4 between the top and bottom surfaces.

[0046] Figure 6 shows (a) an orthographic projection image 52 obtained by XY scanning an actual pottery vessel with the 3D object image scanner 1, and (b) a rectangular image (orthographic projection image) 53 and (c) an annular sector image 54 obtained by rotation scanning.

[0047] The pottery in question is a Jomon pottery with a tapered shape, where the diameter decreases towards the bottom, as shown in the orthographic projection image 52 in Figure 6(a). As shown in Figure 6(b), in the rectangular image 52, it can be seen that the shape of the surface pattern is considerably distorted and expanded laterally from the actual object as you go downwards. In contrast, in the annular sector image 53 in Figure 6(c), the distortion of the surface pattern in the lower half of the bottom surface is considerably reduced, and the pattern is natural and close to the actual object throughout. Thus, the 3D object image scanner 1 can create a full-circumference image with minimal distortion, which is suitable for observing the entire outer circumference of a 3D object.

[0048] Figure 7 shows an example of a system and method for creating and displaying a three-dimensional view of a texture-mapped object using the three-dimensional object image scanner 1 according to the embodiment described above.

[0049] As shown in Figure 7, this three-dimensional viewing system 80 includes a three-dimensional object image scanner 1 according to the embodiment described above, a three-dimensional object shape measuring device 82, and a three-dimensional computer graphics (3DCG) device 84. The three-dimensional object shape measuring device 82 optically measures the object S and creates three-dimensional shape data 86 of the object S. The three-dimensional object shape measuring device 82 is equipped with, for example, a photogrammetry device with a camera, or a distance measuring device such as LiDAR or radar, and creates three-dimensional shape data 86 of the object S (for example, three-dimensional multi-point data or three-dimensional mesh data) by measuring the distance from these devices to numerous points on the outer surface of the object S. Certain models of commercially available mobile phones or personal computers have a built-in three-dimensional object shape measuring device 82, so those models may be used in this system 80.

[0050] The 3DCG device 84 receives three-dimensional shape data 86 of the subject S created by the three-dimensional object shape measurement device 82 and a rectangular image dataset 90 of the subject S created by the three-dimensional object image scanner 1. The 3DCG device 84 applies texture mapping using the rectangular image dataset 90 to the outer surface of the three-dimensional shape data 86 to create a texture-mapped three-dimensional view 88 of the subject S, and displays or outputs it. The 3DCG device 84 may be, for example, a computer equipped with a CPU 130, storage 132, and a display screen 134. It can store a computer program 136 for creating the three-dimensional view 88 in the storage 132, create the three-dimensional view 88 by executing the computer program 136 on the CPU 130, and display the three-dimensional view 88 on the display screen 134. The 3DCG device 84 may be the computer terminal 61 of the three-dimensional object image scanner 1 shown in Figure 1, or another computer terminal.

[0051] Figure 8A shows an example of the data structure of the rectangular image dataset 90 shown in Figure 7.

[0052] As shown in Figure 8A, the rectangular image dataset 90 includes a rectangular image 92 of the portion of the subject S to be photographed (in the illustrated example, the portion to be photographed is the entire subject S, but it may also be a part of the subject S) created by the 3D object image scanner 1 rotating and scanning the subject S in the method already described, and scan condition setting data 94 that shows the setting values ​​of various condition items applied during the photography. The scan condition setting data 94 may include setting values ​​for various condition items set by the user on the scan condition setting screen 10 as shown in Figure 3, such as the starting position, capture height, base position, pottery height, φ base, φ top, Z axis range, scan mode, and layer settings. However, it is not necessary to include setting values ​​for all of these condition items. For example, when using the rectangular image dataset 90 for texture mapping to three-dimensional shape data, as described later, it is sufficient that the scan condition setting data 94 includes data that can identify which range in the height direction of the subject S is captured in which range in the height direction of the rectangular image 92 (for example, the starting position, capture height, base position, and pottery height).

[0053] The rectangular image dataset 90 shown in Figure 8A is an example where the subject S is a coffee cup as shown in Figure 7. In this example, the rectangular image 92 is an orthographic projection image of the appearance (outer surface texture) of the subject S over a range from rotation angle θ=0 degrees (starting angle 70) to θ=360 degrees, obtained by scanning the subject S while rotating it clockwise around the central axis 31 as indicated by the arrow 62. In this example, the rectangular image 92 includes an orthographic projection image 96 of the body 72 of the subject S, which is a coffee cup, and an orthographic projection image 98 of the handle 74.

[0054] The rectangular image dataset 90 shown in Figure 8A has the meaning shown in Figure 8B. As shown in Figure 8B, the dimension (height) of the shorter side of the rectangular image 92 is the capture height y2. The height position of the bottom edge of the rectangular image 92 is the starting position y1, and the height position of its top edge is y3 = y1 + y2. The dimension (height) of the shorter side of the orthographic projection images 96 and 98 of the subject S within the rectangular image 92 is the pottery height y5. The height position of the bottom edge of the orthographic projection images 96 and 98 of the subject S is the base position y4, and the height position of its top edge is y6 = y4 + y5. The left side of the rectangular image 92 corresponds to a rotation angle θ = 0 degrees, i.e., a starting angle of 70, and the right side corresponds to a rotation angle θ = 360 degrees (starting angle 70).

[0055] If we virtually round a rectangular image 92, as shown in Figure 8B, with a rotation angle θ ranging from 0 to 360 degrees, around the central axis 31 and connect the left side (θ=0 degrees) and the right side (θ=0 degrees), the rectangular image 92 becomes a three-dimensional cylinder as shown in Figure 8C. Therefore, the rectangular image dataset 90 shown in Figure 8A is also three-dimensional cylindrical data 90, representing the appearance (outer surface texture) of the part of the subject S that we want to photograph (the entire subject S in the example shown) as a three-dimensional cylindrical body as shown in Figure 8C. Here, in the illustrated example, the rectangular image 92 includes an image of the entire appearance of the subject S, but this is not necessarily required. For example, if the part of the subject S that we want to texture map is only a part of the subject S, then it is sufficient for the rectangular image 92 to contain only an image of that part.

[0056] Figure 9 shows the flow of the process of creating a three-dimensional view of the subject S, performed by the 3DCG device 84 shown in Figure 7.

[0057] As shown in Figure 9, in step 200, the 3DCG device 84 receives three-dimensional shape data 86 of the subject S created by the three-dimensional object shape measurement device 82. The three-dimensional shape data 86 of the subject S represents the three-dimensional shape of the subject S and may be, for example, multipoint data or mesh data that includes the three-dimensional coordinate values ​​of multiple points on the outer surface of the subject S.

[0058] In step 202, the 3DCG device 84 receives a rectangular image dataset 90 of the subject S created by the 3D object image scanner 1. In Figure 9, for ease of understanding, the rectangular image dataset 90 is represented as three-dimensional cylindrical data, but in reality, it is a set of rectangular images 92 and scan condition setting data 94, as shown in Figure 8A.

[0059] In steps 202 and 204, the 3DCG device 84 performs texture mapping on the three-dimensional shape data 86 using a cylindrical mapping method with the rectangular image dataset 90 as the texture image. Although this texture mapping process is actually done mathematically, a simpler, physical explanation is as follows.

[0060] In step 202, the rectangular image dataset (three-dimensional cylindrical data) 90 and the three-dimensional shape data 86 are arranged relative to each other such that the central axis 31 and starting angle 70 of the former correspond to the central axis and starting angle 70 of the latter. Possible methods for matching the central axes and starting angles of the two include comparing the patterns represented by images 96 and 98 of the subject S in the rectangular image dataset (three-dimensional cylindrical data) 90 with the relief patterns of the three-dimensional shape data 86 and performing matching between the two, or having the user specify the central axis and starting angle of the three-dimensional shape data 86 and align them with the central axis 31 and starting angle 70 of the three-dimensional cylindrical data 90, respectively.

[0061] In step 204, the color values ​​of each pixel 300 of the images 96 and 98 of the subject S within the three-dimensional cylindrical data 90 are assigned to the outermost point 304 of the three-dimensional shape data 86, where it intersects with a perpendicular line 302 drawn from each pixel 300 to the central axis 31. The process in step 204 is performed for all pixels of the images 96 and 98 of the subject S.

[0062] As a result, in step 208, a three-dimensional view 88 of the subject S is created by cylindrically mapping images 96 and 98 of the subject S onto the three-dimensional shape data 86 of the subject S. The 3DCG device 84 displays the three-dimensional view 88 on its display screen 134.

[0063] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations.

[0064] Furthermore, the dimensions, shapes, etc., of each component shown in the illustration are not necessarily accurately represented, and may have been modified as appropriate to emphasize the features of this embodiment. [Explanation of Symbols]

[0065] 1: 3D object image scanner, 4: Rotating table, 7: Carriage, 8: Scanner head, 9: Control device, 31: Central axis, 50, 52: Rectangular image, 51, 53: Annular sector image, 70: Starting angle, 80: Three-dimensional view system, 82: 3D object shape measurement device, 84: Three-dimensional computer graphics (3DCG) device, 90: Rectangular image dataset (three-dimensional cylindrical body data), 92: Rectangular image, 94: Scan condition setting data, A: Object space, S: Subject

Claims

1. A rotary table capable of rotating the subject around a predetermined central axis, A scanner head having a telecentric lens system and acquiring an image of the subject space in front of the telecentric lens system, A moving means for moving the scanner head relative to the object space, A stereoscopic object image scanner comprising a rotary table, a scanner head, and a control device that controls the moving means and processes image data of the subject, The control device is With the scanner head positioned such that the central axis is located in front of the telecentric lens system, the subject is rotated around the central axis while acquiring multiple image data with different rotation angles that are sequentially output from the scanner head. From the plurality of image data, pixel data that is in focus on the subject is detected, and the detected pixel data is integrated to create a rectangular image that represents the appearance of the part of the subject that is to be photographed. A three-dimensional object image scanner that converts the aforementioned rectangular image into a ring-shaped sector image based on the top, bottom, and height dimensions of the part of the subject to be photographed.

2. The control device allows the user to set the portion of the subject to be photographed, the top dimension, the bottom dimension, and the height dimension, as described in claim 1.

3. The annular sector image is configured such that the dimensional ratio between the dimension of its outer arc, the dimension of its inner arc, and the dimension of its radial width is equal to the dimensional ratio between the top dimension, the bottom dimension, and the height dimension, as described in claim 1 or claim 2.

4. The control device outputs the rectangular image and the annular sector image, as described in claim 1 or 2, for a three-dimensional object image scanner.

5. The three-dimensional object image scanner according to claim 1 or 2, wherein the rotating table is movable along the shooting direction of the scanner head.

6. In a three-dimensional computer graphics device that creates a three-dimensional view of a subject, A rectangular image dataset is input, comprising: a rectangular image created by the three-dimensional object image scanner described in claim 1, which represents the appearance of the portion of the subject to be photographed, with a rotation angle around the central axis ranging from 0 to 360 degrees; and supplemental data attached to the rectangular image that defines the height range of the portion of the subject to be photographed in the rectangular image; wherein the appearance of the portion of the subject to be photographed is represented as a three-dimensional cylindrical body. Input the three-dimensional shape data of the aforementioned subject, A three-dimensional computer graphics device configured to create a three-dimensional view of an object by performing texture mapping on the input three-dimensional shape data using the input rectangular image dataset via a cylindrical mapping method.

7. In a method for creating a three-dimensional view of a subject, A rectangular image dataset is input, comprising: a rectangular image created by the three-dimensional object image scanner described in claim 1, which represents the appearance of the portion of the subject to be photographed, with a rotation angle around the central axis ranging from 0 to 360 degrees; and supplemental data attached to the rectangular image that defines the height range of the portion of the subject to be photographed in the rectangular image; wherein the appearance of the portion of the subject to be photographed is represented as a three-dimensional cylindrical body. A method for creating a three-dimensional view of an object by performing texture mapping on input three-dimensional shape data using the input rectangular image dataset with a cylindrical mapping method.

Images