JP3138271B2 - 3D display method of object or space - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a three-dimensional display of three-dimensional analysis results and three-dimensional measurement data by a computer, and relates to a method for automatically determining an optimum display color density.

[Conventional technology]

The present invention relates to volume rendering (Volume Renderin).
g (spatial display method)), as prior art, a technique prior to volume rendering and a volume rendering technique prior to the present invention will be described. As an example of a technique prior to volume rendering, consider a case in which a pressure distribution and a flow velocity distribution of a gas flow in a duct are obtained by a computer and displayed graphically. This is a three-dimensional state, but since three-dimensional display was previously difficult, a method of displaying their distribution on an arbitrary cross section of the analysis region has been widely used.

 FIGS. 11 and 12 show examples of the display.

FIG. 11 shows the absolute value of the velocity distribution of the fluid in the rectangular section passage bent in an inverted L-shape in a vertical section parallel to the flow path. (A) shows the situation where the speed is increasing sequentially in the slowest parts (b), (c), (d) and (e). FIG. 12 shows the same velocity distribution in the passage in a section perpendicular to the passage.

In order to understand the velocity distribution three-dimensionally from these figures, a technician had to construct a three-dimensional image from each section in his head.

Next, the volume rendering display is a technique that has been put into practical use with the improvement of functions of a graphics workstation capable of efficiently handling high-resolution graphics. The purpose of volume rendering is to simultaneously and three-dimensionally display the inside and outside of a solid. In other words, not only the outside of the solid, but also the internal solid and the internal hole that are hidden inside, are simultaneously drawn and illustrated to clearly show the spatial positional relationship between the solids.

FIG. 13 is a display example for the same problem as above. Areas where volume rendering is most commonly used are:
It is a display of CT (Computer Tomograph) measurement data in medical treatment. By adjusting the transparency of each part such as meat and bones, the appearance of the human head, skull, brain, and bleeding part are visualized as if looking into a transparent bottle. If the degree of transparency is adjusted, the bleeding part can be displayed particularly densely. Therefore, by using this method, a specific affected part can be displayed three-dimensionally while also showing the positional relationship with the other part.

In addition, the isovalue contours of the analysis results such as the pressure field used in the embodiment of the present invention and the magnetic field strength.
Urfaces) are being put to practical use, for example, isosurface display created by connecting the positions of the same pressure value to contour lines used in maps and the like.

The theory of volume rendering will be described with reference to FIG. Small solid 1 whose display target space is called many voxels
Think that is filled with. Figure 14 shows only one of them. The 2 that has passed through this voxel reaches the pixel 4 on the display screen 3 while passing through other voxels, and the human eye 5 at the viewpoint sees this.

In this model, the colors entering and leaving one voxel at the position of Xi have the following relationship.

C ⁱ _out , λ (U) = C ⁱ _in , λ (U) (1−α (X)) + cλ (X) α (X) (1) where C ⁱ _out , λ (U): C ⁱ _in , λ (U): darkness of light entering the voxel cλ (X): darkness of voxel at position (X) α (X): position (X) X = (x, y, z): position of voxel in space U = (u, v): position of pixel in screen λ = r, g, b (= red, Green, blue): color component of light The intensity of the color exiting the voxel at position Xi is the intensity of the color entering the voxel at the adjacent Xj.

C ⁱ _out , λ (U) = C ^j _in , λ (U) (2) The color density of the voxel exit at Xj is C ^j _out , λ (U) = C ^j _in , λ (U) (1 −α (X)) + cλ (X) α (X) (3-1) = {C ⁱ _in , λ (U) (1−α (X)) + cλ (X) α (X)} (1- α (X)) + cλ (X) α (X) (3-2) This calculation is repeated to determine the color on the display screen when the display reaches the screen.

Here, cλ (x, y, zo) = c _bkg , λ (= background color) α (x, y, zo) = 1 (= completely opaque) It can display like a lump. Normally, a display such as overlapping colored glass is provided by adding a shade display including reflection on the surface of each solid.

As an example of surface treatment, Phong's model is shown below.

(Surface luminance) = kaaIa + ｛kd ・ (L） N) + ks (H ・ N) ⁿ ｝ Is (5) Here, the first term on the right side is ambient light, and the first term in braces on the right side is diffuse reflection. , The second term relates to specular reflection. This formula is calculated for each of the three primary colors of light, R, G, and B.

ka: reflection coefficient for ambient light Ia: ambient light intensity kd: diffuse reflection coefficient ks: specular reflection coefficient Is: incident light intensity of light from the light source L: vector from the surface toward the light source N: normal vector of the surface H : L and the bisection vector of the vector U from the surface to the viewpoint n: the directivity coefficient of specular reflection [Problem to be Solved by the Invention] The volume rendering technique described above is much more Has three-dimensional expression ability. However, even if the transparency is changed only a little, the display changes to a rag and is very sensitive. In other words, the sense of transparency becomes unbalanced due to the three-dimensional object, the inside becomes almost invisible, or a certain three-dimensional object becomes difficult to see.

Even when trying to make a particular solid stand out, this transparency, reflection and shading are difficult.

Therefore, for fixed applications such as a human head, if the display conditions are tuned in advance according to the display purpose, there is no need for subsequent work. However, in the case where the shape of the analysis target or the shape of the isosurface is not determined in advance, as in the case of displaying the analysis result, the user has to adjust each time. However, the adjustment is sensitive as described above,
Each display takes 10 minutes or more (depending on the number of cells), so the adjustment time for obtaining an appropriate display was long, and it could not be used for daily work.

SUMMARY OF THE INVENTION An object of the present invention is to eliminate the above-mentioned disadvantages and eliminate or improve the efficiency of color adjustment by a user.

[Means for solving the problem]

The purpose of the above is to perform the calculation for color setting in advance on sampling points coarser than the final display, use the result to make color adjustment automatically or interactively, and then This is achieved by performing the following processing. That is, the problem of the prior art is to divide a target object or space into a large number of minute three-dimensional voxels, and to draw a color that a ray connecting a viewpoint and a three-dimensional voxel draws on a display screen composed of a large number of display pixels, in a three-dimensional manner. In the three-dimensional display method of an object or space, which determines the transparency of a voxel and the surface luminance of the surface of an object by geometrical optical calculation, the above-described calculation for color determination is performed in advance on a sample cell having a sampling pitch larger than a display pixel. Then, the maximum density value of the color in the sample cell is determined, the ratio of this value to the maximum density value of the display device is calculated, and then the above-described color determination operation is performed on all display pixels. A three-dimensional display method of an object or space, characterized in that the object or space is multiplied by the obtained density ratio and displayed on the screen as a display density. Divide into a number of three-dimensional voxels, and draw the color that the rays connecting the viewpoint and the three-dimensional voxels draw on the display screen consisting of many display pixels, the transparency of the three-dimensional voxels, the diffuse reflection coefficient of the object surface, the specular reflection coefficient, and the rays themselves In a three-dimensional display method of an object or a space calculated and determined based on the light component of a sample cell having a sampling pitch larger than that of a display pixel, the degree of emphasis on a desired display portion is controlled and a color calculation is determined in advance. Find the highest density value of the color inside, calculate the ratio of this value to the highest density value of the display device,
Next, color determination is performed by controlling the degree of emphasis on a desired portion for all display pixels in the same manner as for the sample cells, and the result is multiplied by the previously obtained density ratio, and this is displayed on the screen as the display density. A method for three-dimensionally displaying an object or space, and a display screen in which a target object or space is divided into a number of minute three-dimensional voxels, and a light beam connecting the viewpoint and the three-dimensional voxel includes a number of display pixels The color to draw on,
In a three-dimensional display method of an object or space, which is calculated and determined based on the transparency of a three-dimensional voxel, the diffuse reflection coefficient of the surface of an object, the specular reflection coefficient, and the light component of the light beam itself, a desired display portion has transparency (α ) Is changed, and the above color calculation is determined in advance for a sample cell having a sampling pitch larger than the display pixel, and the highest density value in the sample cell is obtained. Calculating the ratio of the values, then performing the color determination operation on all display pixels, multiplying the result by the previously determined density ratio, and displaying the result as the display density on the screen. It is solved by a three-dimensional display method of space.

[Action]

Generally, in a three-dimensional flow analysis, an electromagnetic field analysis, or the like, the value rarely rises sharply in an extremely small area, and has a spatial distribution having a certain change gradient. Therefore, for example, if the final display is 1,000 × 1,000 (= one million) pixels, if the sampling is performed every 20 pixels, the calculation points are 50 × 50 = 2500 points,
If a million pixel calculation takes 10 minutes, the calculation time per point is the same, so only 1.5 seconds for the sample point.

The figure obtained from the sample points has poor resolution, but can be used sufficiently for color adjustment. With this sample diagram, automatic setting or manual setting of the color tone is possible.

The calculation time is insignificant as described above even if the color is manually set for the sample diagram and redisplayed.

〔Example〕

FIG. 1 shows a flowchart of computer processing according to one embodiment of the present invention. The purpose of this process is to automatically adjust the displayed color range. In other words, the overall color is prevented from being too dark or too light, and the darkest color to be displayed is automatically adjusted to the darkest color of the display (or the density specified by the user).

The flow of FIG. 1 will be described in order. First, in step 1, pixel sizes PXc and PYc for color setting are set. This is the sampling pitch described above, for example, PXc = PYc = 20
It is. FIG. 2 shows an example of sample points. Even with such a coarse sample, sufficient information necessary for color determination can be obtained.

Next, in 2, the darkest color for each of the three components of red, green, and blue
In preparation for obtaining C _max and λ, each is cleared to zero (initial value setting).

Next, in 3, for each sample point, the color cλ
Calculate The calculation of cλ is performed using the above-described equation (1).
This is based on the shaded display shown by (4) and (5). In the processing flow of FIG. 1, this is handled as one subroutine. And the calculation result cλ is
If it is larger than C _max , λ, it is stored as C _max , λ. Depending on the setting of the transparency, a very large numerical value may be obtained as C _max and λ. Therefore, a floating point number is usually used.

By examining all sample points by the above process, finally C
_The values stored in the memory of _max and λ are the darkest intensities at the sample points.

Next, at 4, C _max of each color of red, green, and blue is compared, and the largest value is set as the deepest color tone C * _max . In this embodiment,
Since it was assumed that each color was balanced, only one darkest color was used regardless of the color.

In 5, the ratio of the thickest tone I _max of the display and the C * _max a color correction factor r. In the case of 16.7 million color display, there are 256 color gradations for each of red, green, and blue. (Red is 8bit, Green is 8bi
t and blue represent colors with 8 bits and 24 bits for convenience, 8 bits = 2 ⁸ = 256 gradations, and 256 ³ = 16.7 million colors. ) I _max = 256, or I _max = 220, for example, assuming that there is a little dark place other than the sample point.

The above processing is preprocessing for automatically determining a color tone.

Next, in step 6, the display pixel sizes PX and PY are determined.
Normally, PX = PY = 1, but practical figures can be obtained with 2 and 3.

7, the same “color cλ” as in the previous processing is applied to all the pixels.
Is calculated, and the obtained cλ is multiplied by the color correction rate r to obtain a display gradation. Then, this color is displayed on the screen. When this process is completed, on the screen image of just a good color range that is limited by I _max is displayed.

The aim of the present invention is not to display the image as it is, but to make the user see what he wants to see. The display according to the flow in FIG. 1 will be described using the object in FIG. 3 as an example.

FIG. 3 shows the display object viewed from directly above. At the viewpoint A, the distance of the light rays passing through the objects 11 and 12 is short, and from the viewpoint B it is long. Therefore, when the colors of these two cases are obtained, there is a great difference between them as shown in FIGS. 4A and 4B. If the color is adjusted at the viewpoint A, the object 12 behind the viewpoint B becomes less visible because the color of the object 11 becomes dark.

However, according to the present invention, for adjusting the display color I _max for the maximum tone C * _max, the phenomenon corresponding to the overexposed and underexposed referred to photograph does not occur, that displayed as well middle and back visible Can be.

The time required for the automatic color adjustment is only 1.5 seconds, depending on the sampling point, as compared with the display time of 10 minutes as described above, and does not impose any burden on the user. In the case of automatic adjustment, since all processing is performed internally, no operation is required by the user.

Further, as shown in the flow of FIG. 1, since the calculation of the color cλ can be performed by using a conventional program as it is, the amount of programs newly developed for automatic adjustment is extremely small.

When the same correction factor is used for the three components of red, green, and blue, the balance of the color tone given to the original object is not lost.

Next, another embodiment for highlighting and displaying a specific object will be described.

The determination of the color includes a transparency and a surface color for displaying a shadow, and the transparency will be described here. Other parameters can be handled similarly.

First, the transparency is set to an initial value for all objects as shown in FIG. The initial values are stored in a database in advance or input by the user. The initial value of the emphasis coefficient is 1.

Here, αλ, i: Transparency for display α ^o λ, i: Initial setting value of transparency for display Eλ, i: Enhancement coefficient i: Object number (i = 1 to i _max ) However, in this embodiment, the color tone (color density) of the object to be viewed is
Is controlled by the emphasis coefficient. Next, the user inputs an enhancement coefficient for each object. If there is no input, the default setting
Leave it at 1.0.

As shown in FIG. 6, the transparency for display calculation is obtained by αλ, i ← Eλ, i × αλ, i (6) for all objects. Eλ, i may include a large value for an object to be emphasized or a small value for an object not to be emphasized. Α set in Expression 6 becomes α used in Expression (1).

By using this αλ, i and displaying according to the flow of FIG. 1, a display in which a specific object is emphasized can be obtained. FIG. 10 shows processing steps for three-dimensionally displaying a target object by using the processes shown in FIGS. 1, 5, and 6. The setting shown in FIG. 5 is executed only once at the beginning, and the processing shown in FIGS. 1 and 6 can be executed any number of times by inputting a command. However, as can be seen from the processing flow of FIG. 6th executed at the time
The processing results shown in the figure are used for the processing shown in FIG. This use example will be described with reference to FIG. FIG. 7 is a plan view in the case where an overlap of three objects is displayed. 8A to 8C show examples of the color tone and the screen display at this time.

FIG. 8A shows a case where all three objects have the same density. All three parts look balanced.

FIG. 8B shows a case where the object 112 is emphasized. In this case, the maximum color tone C _max is larger than the previous example,
It is regulated by the processing flow of FIG. 1 in accordance with the I _max.
In this display, the object 111 is pale, the next object 112 is clearly visible, and the object 113 behind is faintly visible.

FIG. 8C shows a case where the object 113 is emphasized. Objects 111 and 11
Despite 2 being in front, they are pale because object 113
Is clearly visible.

As still another embodiment, FIG. 9 shows an example of an operation screen.

The left side of the screen is for display 21 and the right side is for adjustment. Eight small screens are cut into the adjustment area, seven of them (22 to 28) are parts having different colors, and the other is a sample screen 29 for voxel rendering. Since these eight small screens are displayed corresponding to the sample points, the processing time is short. By providing a slide-type display of a horizontal bar for each part, picking it with a mouse and expanding and contracting it, the density of the display can be adjusted. That is, the value of the slider is set to the value of Eλ, i in the above equation (6). In this case, however, the same emphasis degree Er, i = Eg, i = Eb, i (r, g, b = red, green, blue) is used for each color. The result of this adjustment can be efficiently displayed on the sample screen.

Further, a slider 30 for controlling a color tone range of three primary colors of red, green and blue is provided at the lower right. This allows the color balance to be adjusted. For example, if you want to make red objects stand out,
As shown in the figure, the red range may be a bright range and the others may be a dark range.

 Here, the display color is calculated as follows.

Here, Iλ: display color I _min , λ: minimum range set by the slider I _max , λ: maximum range set by the slider C _max , λ: _maximum color tone cλ: color tone of the pixel to be displayed.

When the colors can be adjusted as described above, the details are displayed on the left display screen.

〔The invention's effect〕

According to the present invention, it is possible to easily adjust the display of voxel rendering that is sensitive to a change in display conditions, which was conventionally time-consuming and very difficult.

That is, the range of the display color that differs depending on the transparency and the thickness of the object can be automatically set by examining the sample points.

For objects that want to stand out, just enter the degree of emphasis,
It can be displayed by the automatic range setting. Further, by adding an interactive operation using the graphic display, the user's sensitivity can be easily reflected.

Since the above adjustment uses sample points, the execution time is short and can be incorporated into the interactive processing. In addition, the amount of development programs is small and easy to incorporate.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a system diagram of color tone processing by a computer according to an embodiment of the present invention, FIG. 2 is an explanatory diagram of how to take sample points for color tone adjustment in the present invention, and FIG. 4A and 4B are diagrams showing color tones at two viewpoints in FIG. 3, and FIG. 5 is a flow chart for initial setting of transparency in the present invention. FIG. 6, FIG. 6 is a flow chart for setting the transparency using the input degree of emphasis in the present invention, FIG.
8A, 8B, and 8C are explanatory diagrams of the color tone and image color display according to the degree of emphasis in the case of FIG. 7, and FIG. 9 is a configuration of a color setting interactive processing screen according to another embodiment of the present invention. FIG. 10 is an explanatory diagram showing processing steps for three-dimensionally displaying a target object using the processes shown in FIG. 1, FIG. 5 and FIG.
FIG. 12 is an example of a conventional sectional display for a three-dimensional object.
FIG. 13 is a diagram showing an example of a three-dimensional display of the object shown in FIGS. 11 and 12, and FIG. 14 is an explanatory diagram of a conventional volume rendering technique. 1 ... Color setting pixel (sample cell) size setting step, 2 ... Color density initial value setting step, 3 ... Calculation of color cλ for all color setting pixels (sample cell), C
_max , λ calculation step, 4... the darkest color tone determination step for all red, green, and blue color components, 5... maximum density correction rate calculation step, 6... display pixel size setting step, full display Calculation and display of colors for pixels.

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G06T 15/00 100

Claims (5)

(57) [Claims] An object or space to be processed is divided into a number of minute three-dimensional voxels, and a color connecting a viewpoint and a three-dimensional voxel is drawn on a display screen composed of a number of display pixels, and the transparency of the three-dimensional voxel is determined. In a three-dimensional display method of an object or space in which the surface luminance of the surface of an object is determined by geometrical optics, the above-described color determination operation is performed in advance on a sample cell having a sampling pitch larger than a display pixel, Find the highest density value of the color in the cell, calculate the ratio of this value to the highest density value of the display device,
Next, the color determination operation is performed on all display pixels, the result is multiplied by the previously determined density ratio, and the result is displayed on the screen as the display density. Method. 2. A method of dividing a target object or space into a number of minute three-dimensional voxels, and defining a color in which light rays connecting the viewpoint and the three-dimensional voxels are drawn on a display screen including a large number of display pixels, the transparency of the three-dimensional voxels. , The diffuse reflection coefficient of the object surface,
In a three-dimensional display method of an object or a space which is calculated and determined based on a specular reflection coefficient and a light component of a light beam itself, a color calculation is performed in advance by controlling the degree of emphasis on a desired display portion for a sample cell having a sampling pitch larger than a display pixel. A determination is made to determine the highest density value of the color in the sample cell, calculate the ratio of this value to the highest density value of the display device, and then determine the degree of enhancement for the desired portion for all display pixels as well as for the sample cell. Control to determine the color, multiply the result by the density ratio obtained earlier,
A three-dimensional display method of an object or a space, which displays this on a screen as a display density. 3. A method of dividing a target object or space into a large number of minute three-dimensional voxels, and drawing a color connecting a viewpoint and a three-dimensional voxel on a display screen composed of a large number of display pixels to a transparency of the three-dimensional voxel. , The diffuse reflection coefficient of the object surface,
In a three-dimensional display method of an object or a space calculated and determined based on a specular reflection coefficient and a light component of a light ray itself, a color calculation is performed by changing transparency (α) with respect to a desired display portion from other portions, For the sample cell having a sampling pitch larger than the display pixel, the above color calculation is determined in advance, the highest density value in the sample cell is obtained, the ratio of this value to the highest density value of the display device is calculated, and then all display pixels are processed. A three-dimensional display method of an object or a space, wherein the color determination operation is performed on the result, the result is multiplied by the previously determined density ratio, and the result is displayed on a screen as a display density. 4. The method according to claim 3, wherein the object or space to be processed is divided and displayed in a plurality of parts, and transparency is adjusted for each divided display.
A three-dimensional display method of an object or a space, wherein a color is calculated and displayed for sample points for all portions after the transparency is adjusted. 5. The display range (minimum value, maximum value) for each of the red, green, and blue color components according to claim 1, 2, 3, or 4, and the display range. A three-dimensional display method of an object or a space, wherein a result of a color operation is assigned to a sample point and displayed.