JP5354494B2 - 3D image generation apparatus, 3D image generation method, and program - Google Patents

Description

  The present invention relates to a three-dimensional image generation apparatus that generates a three-dimensional image from a two-dimensional image, a three-dimensional image generation method, and a program that causes a computer to execute the method.

  Conventionally, an image captured by an endoscope has been used in the medical field or the like. The image captured by the endoscope has a narrow field of view because the viewpoint is fixed.

  On the other hand, in clinical sites such as respiratory organs, digestive organs, and brain surgery, three-dimensional images (referred to as “virtual endoscopic images”) are generated using continuous images acquired by CT or MRI. The virtual endoscopic image has an advantage that the viewpoint is free and the blind spot can be invalidated.

International Publication WO2007 / 139187 Pamphlet

  Since the resolution of the virtual endoscopic image depends on the spatial resolution of CT or MRI, the virtual endoscopic image generated from CT or MRI is inferior to the actual image. Therefore, for example, even if a virtual endoscopic image is viewed, a lesion of 5 mm or less cannot be detected. In the case of a virtual endoscopic image generated from CT or MRI, there is no color information. Therefore, it is difficult to diagnose intraepithelial cancer and invasive cancer using, for example, virtual endoscopic images generated from CT or MRI. As described above, the conventional virtual endoscopic image has a problem of low usefulness because it has low resolution and no color information.

  The present invention solves the above-described problem, and is a three-dimensional image generation apparatus that generates a highly useful three-dimensional image (for example, a virtual endoscopic image) having high resolution and color information. An object is to provide an image generation method and a program for causing a computer to execute the method.

  A second 3D image generation apparatus according to the present invention includes an input unit that inputs a 2D image of a subject having a 3D shape, a storage unit that stores the input 2D image, and a 2D image from the storage unit. And a control means for generating a three-dimensional image of the subject from the read two-dimensional image. The control means obtains a difference between the blue luminance information and the green luminance information for each pixel in the two-dimensional image, calculates a coordinate on the third coordinate axis of each pixel based on the difference, and obtains the three-dimensional coordinate. A three-dimensional image of the subject is generated based on the three-dimensional coordinates.

  The second three-dimensional image generation method according to the present invention is a method for generating a three-dimensional image from a two-dimensional image of a subject having a three-dimensional shape using an information processing apparatus. In the method, the control means of the information processing apparatus obtains a difference between the blue luminance information and the green luminance information for each pixel in the two-dimensional image obtained by imaging the subject, and based on the difference, the third coordinate axis of each pixel. Calculating upper coordinates to obtain three-dimensional coordinates; and generating a three-dimensional image of the subject based on the three-dimensional coordinates of each pixel in the two-dimensional image.

  According to the present invention, three-dimensional coordinates are calculated using luminance information of a two-dimensional image, a three-dimensional model is generated based on the calculated three-dimensional coordinates, and a texture of the two-dimensional image is pasted on the three-dimensional model. Attached. Therefore, the resolution and color information of the two-dimensional image are applied to the three-dimensional image as they are. Specifically, for example, resolution and color information obtained from an image captured using an endoscope is applied to a three-dimensional image as it is. Therefore, a highly useful three-dimensional image having high resolution and color information can be generated.

The block diagram which shows the structure of the three-dimensional image generation apparatus of Embodiment 1 and 2 of this invention. The figure for demonstrating the method to produce | generate a two-dimensional image by expand | deploying planarly the endoscopic image obtained by image | photographing a tubular structure in Embodiment 1 of this invention. 3 is a flowchart showing a method for generating a three-dimensional image by the three-dimensional image generating apparatus according to the first and second embodiments of the present invention. FIG. 3 is a schematic diagram illustrating conversion from a two-dimensional image to a three-dimensional image in the first embodiment performed using the method of FIG. 3. The flowchart which shows the calculation method of the three-dimensional coordinate in Embodiment 1 of this invention. FIG. 5 is a diagram for explaining elements used when obtaining three-dimensional coordinates. The figure which shows the example of the two-dimensional image used or the produced | generated three-dimensional image in execution of the method shown in Embodiment 1 of this invention. The flowchart which shows the calculation method of the three-dimensional coordinate in Embodiment 2 of this invention. The figure which shows the example of the two-dimensional image used in execution of the method shown in Embodiment 2 of this invention, or the produced | generated three-dimensional image.

Embodiment 1
The three-dimensional image generation apparatus according to the first embodiment of the present invention generates a three-dimensional image from a two-dimensional image. Specifically, in the present embodiment, a two-dimensional image is generated by planarly developing a video imaged using an endoscope (referred to as an “endoscopic image”), and the two-dimensional image is generated. A virtual endoscopic image (three-dimensional image) that is an image obtained by reproducing the original endoscopic image is generated using the image. More specifically, in this embodiment, the shape of the subject of the two-dimensional image is known in advance, and the three-dimensional shape is restored using a calculation formula corresponding to the subject.

1. Configuration FIG. 1 shows a configuration of a three-dimensional image generation apparatus according to Embodiment 1 of the present invention. The three-dimensional image generation apparatus according to the present embodiment generates an endoscope apparatus 200 that generates endoscope image data from an image captured by an endoscope scope, and generates a two-dimensional developed image from the endoscope image data. And an information processing apparatus 100 that generates a three-dimensional image from the two-dimensional developed image.

  The information processing apparatus 100 is, for example, a personal computer. The information processing apparatus 100 receives the image data from the endoscope apparatus 200, generates a two-dimensional developed image, and generates an information processing unit 11 that generates a three-dimensional image from the two-dimensional developed image, and the two-dimensional image or the three-dimensional image. Display unit (for example, display) 12 and an operation unit (mouse 13a and keyboard 13b) for inputting commands to the information processing unit 11 and the display unit 12 from the user.

  The information processing section 11 stores an input / output interface (I / F) 11a that externally transmits and receives 2D images and 3D image data, and 3D images generated by converting 2D images and 2D images. And a control unit 11c that generates a three-dimensional image from the two-dimensional image stored in the storage unit 11b and the two-dimensional image read from the outside via the input / output interface 11a.

  The endoscope apparatus 200 is inserted into a tubular structure to control the endoscope scope 21 that images the inner surface of the tube, and signals input via the endoscope scope 21. And a control unit 22 for creating a video file (endoscopic image) based on the control unit 22. The endoscope apparatus 200 is a medical device that incorporates an optical system for observing a cavity in the body.

  The control unit 22 includes a control unit 22a that controls each component in the unit 22, a signal processing unit 22b that creates a video file on the inner surface of the tubular structure based on a signal input via the endoscope scope 21, and A light source unit 22c that is a light source of illumination light emitted from the distal end portion of the endoscope scope 21 to the observation target. The control unit 22a controls imaging and lighting on / off switching by the endoscope scope 21 and adjusts the amount of light supplied from the light source unit 22c to the endoscope scope 21 in accordance with a user operation.

  The endoscope scope 21 includes an objective lens, a pair of illumination optical fibers, and an operation channel such as a forceps channel and a suction channel, so that they are respectively exposed at the distal end portion of the endoscope scope 21. Is provided. Such a configuration is well known in the art, and the endoscope scope 21 is not limited to this configuration. For example, a configuration in which a single or three or more illumination optical fibers are provided is adopted. May be.

  The information processing apparatus 100 and the endoscope apparatus 200 configured as described above cooperate to perform the operation described in the image construction technique disclosed in the international publication WO2007 / 139187 pamphlet, and thereby the inner surface of the tubular structure is formed. An image is taken and a two-dimensional image composed of a plurality of frame images is generated.

  The information processing apparatus 100 and the endoscope apparatus 200 are connected via a cable 50 such as a USB cable. Between the information processing apparatus 100 and the endoscope apparatus 200, a two-dimensional image generated by the endoscope apparatus 200 is transmitted to the information processing apparatus 100, or a command signal is transmitted from the information processing apparatus 100 to the endoscope apparatus 200. Can be transmitted. A video capture board may be interposed between the information processing apparatus 100 and the endoscope apparatus 200.

  Note that, in the present embodiment, a case where a three-dimensional image is generated from an endoscopic image photographed by the endoscope apparatus 200 is illustrated, and thus the three-dimensional image generation apparatus includes the endoscope apparatus 200. However, since the generation of a three-dimensional image from a two-dimensional image can be executed only by the information processing apparatus 100, the three-dimensional image generation apparatus does not necessarily include the endoscope apparatus 200.

2. Operation 2-1.2 Generation of Two-Dimensional Image The processing for generating a two-dimensional developed image by the information processing apparatus 100 uses the image construction technique disclosed in the international publication WO2007 / 139187 pamphlet. This image construction technique will be described. The information processing apparatus 100 acquires an endoscopic image (moving image or a plurality of still images) of the inner wall of the tubular structure captured by the endoscope scope 21 from the endoscope apparatus 200, and planarly displays the inner wall of the tubular structure. A developed two-dimensional developed image is generated. Specifically, as shown in FIG. 2A, a test line R is set on each of the images F1, F2, F3,... Constituting a moving image or a plurality of still images, and on the test line R Only pixels are extracted, and pixel column data images R1, R2, R3,... Are generated by one-dimensionally developing the extracted pixels for each image. Thereafter, as shown in FIG. 2B, the inner wall of the tubular structure is planarized by synthesizing a plurality of consecutive pixel row data images R1, R2, R3,. A developed two-dimensional image P is generated. FIG. 2C is a diagram showing an example of an image P that is generated from a continuous endoscopic image as shown in FIG. Thus, the international publication WO2007 / 139187 pamphlet discloses a technique for generating a two-dimensional image P in which a tubular structure inner wall is developed in a plane. The information processing apparatus 100 of this embodiment has the above function.

  The two-dimensional image (two-dimensional developed image) generated by the endoscope apparatus 200 is transmitted to the information processing apparatus 100 and stored in the storage unit 11b.

2-2.2 Conversion from a Two-Dimensional Image to a Three-Dimensional Image A method for generating a three-dimensional image from a two-dimensional image will be described with reference to FIGS. 3 and 4. FIG. 3 is a flowchart illustrating a method for generating a three-dimensional image by the control unit 11 c of the information processing apparatus 100. FIG. 4 is a diagram for explaining a conversion method from a two-dimensional image to a three-dimensional image when the subject of the two-dimensional image is a tubular structure. Referring to FIG. 3, first, for the generated three-dimensional image (FIG. 4B), the number of vertices of the two-dimensional image (ie, the number of vertices specified by the user is set in the control unit 11 c (that is, The control unit 11c for setting (spatial resolution) inputs the set number of vertices (S301). Here, the number of divisions in the horizontal direction (U-axis direction in FIG. 4A) and the vertical direction (V-axis direction in FIG. 4A) of the two-dimensional image are set simultaneously. Based on the set number of vertices (number of divisions), the control unit 11c divides the two-dimensional image into a lattice shape as shown in FIG. 4A (S302). The control unit 11c sets the two-dimensional coordinates (u, v) (texture coordinates) of the two-dimensional image divided in a lattice shape (S303). The control unit 11c calculates luminance information on each coordinate in the two-dimensional image, and calculates three-dimensional coordinates (x, y, z) (model space coordinates) based on the luminance information (S304). The control unit 11c generates a three-dimensional model based on the three-dimensional coordinates (S305). As shown in FIG. 4 (b), the “three-dimensional model” is an object generated on a virtual three-dimensional space having a configuration (TRIANGLE-STRIP) in which triangles composed of three adjacent vertices are continuous. is there. The control unit 11c pastes the texture of the two-dimensional image on the three-dimensional model based on the two-dimensional coordinates and the three-dimensional coordinates (S306). 4A and 4B, the texture of the two-dimensional image is pasted on the three-dimensional model by assigning the U axis to the Z direction and the V axis to the circumferential direction of the XY plane.

  As described above, in the present embodiment, a three-dimensional image is generated from a two-dimensional image based on the luminance information of the two-dimensional image.

  Next, the step (S304) of calculating the three-dimensional coordinates from the luminance information of the two-dimensional image will be described in more detail with reference to FIGS. FIG. 5 is a flowchart illustrating a method for calculating three-dimensional coordinates based on luminance information of a two-dimensional image, which is performed by the control unit 11c of the information processing apparatus 100. FIG. 6 is a diagram for explaining elements (arguments) used when obtaining three-dimensional coordinates. In the three-dimensional coordinate calculation method according to FIG. 5 and the explanatory diagram of FIG. 6, when the subject of the two-dimensional image from which the three-dimensional image is generated is a tubular structure, that is, by the endoscope apparatus 200, a lumen or the like. This shows a case where an image having a tubular shape is photographed and a two-dimensional image generated from the photographed image is used.

Referring to FIG. 5, the control unit 11c of the information processing apparatus 100 calculates the luminance value L of each coordinate of the two-dimensional image (S501). Specifically, the luminance value L is calculated by the following formula (1) using a weighted average method using NTSC coefficients.
L = 0.298912 × R + 0.586611 × G + 0.114478 × B (1)
(R: red luminance value, G: green luminance value, B: blue luminance value)

  Based on the calculated luminance value L, the control unit 11c calculates a distance R ′ from the central axis of the endoscope scope 21 to the observation point (tissue) (S502). At this time, as shown in FIG. 6, the distance R ′ from the central axis of the endoscope scope 21 to the observation point is the observation point from the tip of the optical fiber of the endoscope scope 21 (hereinafter also referred to as “light source”). Is proportional to the distance R. Therefore, the distance R from the tip of the optical fiber of the endoscope scope 21 to the observation point is calculated using the luminance value L of the vertex, and the calculated distance R is measured from the central axis of the endoscope scope 21 to the observation point. Is used as the value of the distance R ′. Specifically, the following calculation is performed.

（ｋ_ｄ：物体の拡散反射率、Ｉ_ｑ：光源光度、θ：光線の入射角、γ：表示部のガンマ特性）
上記式（２）で示すように、本実施形態においては、内視鏡装置２００のガンマ特性を考慮して、非線形な信号を線形な信号に補正するために、２次元画像の各画素の輝度値に対して、ガンマ補正を行っている。 The relationship between the luminance value L of the vertex and the distance R from the tip of the optical fiber of the endoscope scope 21 to the observation point is expressed by the following formula (2).

(K _d : diffuse reflectance of object, I _q : luminous intensity of light source, θ: incident angle of light, γ: gamma characteristic of display unit)
As shown by the above equation (2), in this embodiment, in order to correct a nonlinear signal to a linear signal in consideration of the gamma characteristics of the endoscope apparatus 200, the luminance of each pixel of the two-dimensional image is corrected. Gamma correction is performed on the value.

In the above formula (2), since the light source luminous intensity I _q , the diffuse reflectance k _{d of the} object, and the incident angle θ of the light ray are unknown, assuming that these values are constant, the center of the endoscope scope 21 The distance R ′ from the axis to the observation point is expressed by the following equation (3). A distance R ′ from the central axis of the endoscope scope 21 to the observation point is obtained by Expression (3).

  The range of values that the luminance value L can take is “0.0 to 1.0”. When the distance R ′ is obtained by the above equation (3), the range that the distance R ′ can take is “1.0 to ∞”. This is contrary to the fact that the distance R from the light source of the endoscope scope to the observation point is 0.0 or more. Therefore, after obtaining the distance R ′ by the above equation (3), the control unit 11c calculates the relative distance R ″ by correcting the distance R ′ (S503). It calculates | requires by performing the following processes (1) -process (3).

Process 1: “R ′” is shifted to “R′−1.0”.
(R ′ can take a range of “0.0 to ∞”)
Process 2: “R′−1.0” is divided by the maximum value of the entire R ′ value.
(R ′ can take a range of “0.0 to 1.0”)
Process 3: Compression is performed within a range of “predetermined minimum value min (> 0.0) to 1.0”.
(Range of R ′: “min to 1.0”)

  Thus, first, in order to correct the point corresponding to the luminance value L = 1.0 (the point closest to the endoscope) to a point at a distance “0” from the light source, the entire set of distance R ′ values is set to 1. .0 shift (process 1). As a result, the range that R ′ can take becomes “0.0 to ∞”. Next, each value of the distance R ′ is divided by the maximum value of the entire R ′ value and normalized (process 2). Thereby, the range which distance R 'can take becomes "0.0-1.0". In the normalized value R ′, the point closest to the optical axis is “0”, that is, the object exists at the same position as the optical axis. However, since the endoscope scope 21 has a thickness, it is impossible for an object to exist at the same position as the optical axis (that is, an object enters the endoscope scope diameter). Therefore, based on the ratio between the target tissue and the endoscope scope diameter, the central portion of the lumen is set as the scope progress region, and the point of distance “0” is excluded from the range that R ′ can take. Thereby, the range that R ′ can take becomes a predetermined minimum value min (> 0.0) to 1.0 ”. The distance R ′ corrected in this way is defined as a relative distance R ″. In this embodiment, the distances R, R ′, and R ″ all indicate relative distances rather than absolute values.

  Next, the control unit 11c calculates the angle rad from the number of vertices in the circumferential direction (circumferential spatial resolution) (S504). Thereafter, using the predetermined weight w1, the control unit 11c calculates the x coordinate by “R” × w1 × sin (rad) ”(S505). Further, the y-coordinate is calculated by “R” × w2 × cos (rad) ”using the predetermined weight w2 (S506). Here, the weights w1 and w2 used in the calculation of the x coordinate and the calculation of the y coordinate are values set arbitrarily. The weights w1 and w2 may be the same value. Finally, the z coordinate is calculated from the number of vertices in the length direction (length direction spatial resolution) (S507). The controller 11c determines whether or not the three-dimensional coordinates (x, y, z) have been calculated for all vertices (S508). If the three-dimensional coordinates have not been calculated for all the vertices, the process returns to step S501, and the above processing S501 to 507 is repeated until the three-dimensional coordinates are calculated for all the vertices.

  A three-dimensional model is generated based on the three-dimensional coordinates obtained as described above, and a texture is pasted.

  FIG. 7 shows an example of a three-dimensional image obtained by the three-dimensional image generation apparatus of this embodiment. FIG. 7A shows an image obtained by imaging the human large intestine with the endoscope scope 21. FIG. 7B shows a two-dimensional developed image generated by the information processing apparatus 100. FIG. 7C shows a three-dimensional image (virtual endoscopic image) generated from the two-dimensional developed image by the information processing apparatus 100. FIG. 7D shows an example in which the virtual endoscopic image is rotated and displayed (an image in which the left side surface of the lumen is observed from the right direction). FIG. 7D shows a case in which the front wall is automatically erased (not displayed) by turning on the setting of the culling mode, and only the rear wall surface is displayed. Here, the “culling mode” is a mode in which the front / back information of each polygon in the three-dimensional model is read, and only the polygon showing the surface on the specific side with respect to the camera is drawn. In the three-dimensional model generated by the information processing apparatus 100 according to this embodiment, each polygon defines a surface facing the central axis of the lumen as the front side. By turning on the “culling mode” under the setting to draw only the front surface, the polygon showing the back side can be erased and the rear wall surface can be drawn. The three-dimensional image generated by the information processing apparatus 100 according to the present embodiment can be arbitrarily rotated or an arbitrary cross section can be displayed.

3. Summary According to the present embodiment, since the three-dimensional image is generated using the texture of the two-dimensional image obtained from the endoscope image, the color information of the original texture can be obtained and high resolution can be obtained. Therefore, the three-dimensional image generated by the present embodiment is highly useful. Therefore, it is possible to grasp the properties of the entire tissue, the positional relationship between the three-dimensional tissues, and the like using the three-dimensional image generated according to the present embodiment. Moreover, the accuracy of detecting a lesion is improved by displaying an image with an arbitrary cross section or viewing the whole while arbitrarily rotating the target organ.

  Further, according to the present embodiment, a three-dimensional image in which the original shape is restored is generated using luminance information. That is, when the shape of the subject of the two-dimensional image is known in advance, the third coordinate (z-axis coordinate) is calculated using the calculation formula based on the luminance information, and the three-dimensional shape is restored. be able to. For example, as in this embodiment, a three-dimensional image (virtual endoscopic image) can be generated directly from a real image (endoscopic image) captured by the endoscope scope 21. In the actual image captured by the endoscope scope 21, the viewpoint is fixed, the field of view is narrow, and it is difficult to specify the location of the region of interest. However, according to the present embodiment, since a three-dimensional image with free viewpoint can be generated, the field of view is widened and it is easy to specify the location of the region of interest. Thereby, a doctor or the like can observe the region of interest from a free viewpoint.

  In this embodiment, the three-dimensional coordinate calculation formula “(x, y, z) = (R ″ × w1 × sin (rad), R ″ × w2 × cos” shown in steps (S505 to S507) in FIG. (Rad), z) "is applied to a tubular shape. Therefore, in the case of other shapes, the three-dimensional coordinates may be calculated using other calculation formulas. That is, when the shape of the subject of the two-dimensional image is known in advance, the three-dimensional image can be generated by calculating the three-dimensional coordinates by using a calculation formula corresponding to the shape of the subject.

<< Embodiment 2 >>
In the present embodiment, another method of calculating three-dimensional coordinates (S304) will be described. The configuration of this embodiment is the same as that of the first embodiment. In the present embodiment, a case where the z coordinate is calculated based on the luminance information of the image even when the shape of the subject of the two-dimensional image is not known will be described. Specifically, a three-dimensional image is generated from a two-dimensional image by giving a depth (z direction) to the two-dimensional image. More specifically, when calculating the three-dimensional coordinates (x, y, z), the values of the original two-dimensional coordinates (u, v) are used for the x-coordinate and y-coordinate, and only the z-coordinate is luminance. Calculate from the value. In the present embodiment, the configuration of the three-dimensional image generation apparatus is the same as that in the first embodiment, but the endoscope image 200 may not be provided.

  FIG. 8 shows a flowchart of the calculation of the three-dimensional coordinates in this embodiment. The control unit 11c of the information processing apparatus 100 calculates the blue luminance value of the pixel at the vertex of the two-dimensional image (S801). The control unit 11c calculates the green luminance value of the pixel at the vertex of the two-dimensional image (S802). The control unit 11c assigns the two-dimensional coordinate values (u, v) to the x- and y-coordinates of the three-dimensional coordinates, and sets the z-coordinate of the pixel at the vertex based on the blue luminance value and the green luminance value. Calculate (S803). Specifically, the z-coordinate is calculated so that the z-axis coordinate becomes larger (or smaller) as the difference value obtained by subtracting the green luminance value from the blue luminance value is larger. The control unit 11c determines whether or not the three-dimensional coordinates (x, y, z) have been calculated for all vertices (S804). Steps S801 to S803 are repeated until the three-dimensional coordinates for all vertices are calculated.

  FIG. 9 shows an example in which a texture is pasted on the three-dimensional coordinates obtained as described above according to the flow of FIG. FIG. 9 shows a result of an experiment performed on the fundus oculi when a three-dimensional image is generated by calculating the z-coordinate using the method of FIG. 9A shows an original two-dimensional image, FIG. 9B shows an image converted into three dimensions by the image generation method of the present embodiment, and FIGS. 9C to 9E show FIGS. It is a figure which shows the state which rotated the three-dimensional image of b) and cut | disconnected by arbitrary cross sections. It was confirmed that the z coordinate closer to the actual value can be obtained by subtracting the green luminance value from the blue luminance value from the image created by the image generation method of the present embodiment.

  According to this embodiment, a three-dimensional image can be generated regardless of the shape of the subject of the two-dimensional image. At this time, in particular, the accuracy of the three-dimensional image is improved by calculating the z-coordinate based on the difference between the green luminance value and the blue luminance value.

  In addition, you may make a computer perform the generation method of the three-dimensional image of the said Embodiment 1, 2 by a program.

  In addition, this invention can be used not only for a medical use but for industrial uses, such as a test | inspection of the cracks of piping, such as a clay pipe, various piping, and an engine, and its versatility as an inspection apparatus is wide.

  The present invention has an effect that a highly useful three-dimensional image having high resolution and color information can be generated, and is useful for medical use, industrial use, and the like.

DESCRIPTION OF SYMBOLS 11 Information processing part 11a Input / output interface 11b Storage part 11c Control part 12 Display part 13 Operation part 13a Mouse 13b Keyboard 21 Endoscope 22 Control unit 22a Control part 22b Signal processing part 22c Light source part 100 Information processing apparatus 200 Endoscope apparatus

Claims (6)

Input means for inputting a two-dimensional image of a subject having a three-dimensional shape;
Storage means for storing the input two-dimensional image;
Control means for reading the two-dimensional image from the storage means and generating a three-dimensional image of the subject from the read two-dimensional image;
The control means obtains a difference between the blue luminance information and the green luminance information for each pixel in the two-dimensional image, calculates a coordinate on the third coordinate axis of each pixel based on the difference, and calculates the three-dimensional Obtaining coordinates and generating a three-dimensional image of the subject based on the three-dimensional coordinates;
3D image generation device. The 2-dimensional image is an image generated by developing the image captured by the endoscope in a plane, the three-dimensional image generating apparatus of the mounting serial to claim 1. A method of generating a 3D image from a 2D image of a subject having a 3D shape using an information processing device,
The control means of the information processing apparatus
For each pixel in the two-dimensional image in which the subject is imaged, the difference between the blue luminance information and the green luminance information is obtained, and the coordinates on the third coordinate axis of each pixel are calculated based on the difference to obtain the three-dimensional coordinates. A step of seeking
Generating a three-dimensional image of the subject based on a three-dimensional coordinate of each pixel in the two-dimensional image.
3D image generation method. The three-dimensional image generation method according to claim 3, wherein the two-dimensional image is an image generated by planarly developing an image photographed by an endoscope. A program for causing an information processing device to generate a three-dimensional image from a two-dimensional image of a subject having a three-dimensional shape,
In the control means of the information processing apparatus,
For each pixel in the two-dimensional image in which the subject is imaged, the difference between the blue luminance information and the green luminance information is obtained, and the coordinates on the third coordinate axis of each pixel are calculated based on the difference to obtain the three-dimensional coordinates. A step of seeking
Generating a three-dimensional image of the subject based on the three-dimensional coordinates of each pixel in the two-dimensional image. The three-dimensional image generation program according to claim 5, wherein the two-dimensional image is an image generated by expanding an image captured by an endoscope in a planar manner.

Images